[{"abstract":[{"text":"Intensive care units (ICU) are increasingly looking towards machine learning for methods to provide online monitoring of critically ill patients. In machine learning, online monitoring is often formulated as a supervised learning problem. Recently, contrastive learning approaches have demonstrated promising improvements over competitive supervised benchmarks. These methods rely on well-understood data augmentation techniques developed for image data which do not apply to online monitoring. In this work, we overcome this limitation by\r\nsupplementing time-series data augmentation techniques with a novel contrastive\r\nlearning objective which we call neighborhood contrastive learning (NCL). Our objective explicitly groups together contiguous time segments from each patient while maintaining state-specific information. Our experiments demonstrate a marked improvement over existing work applying contrastive methods to medical time-series.","lang":"eng"}],"author":[{"last_name":"Yèche","full_name":"Yèche, Hugo","first_name":"Hugo"},{"full_name":"Dresdner, Gideon","last_name":"Dresdner","first_name":"Gideon"},{"first_name":"Francesco","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Matthias","full_name":"Hüser, Matthias","last_name":"Hüser"},{"first_name":"Gunnar","full_name":"Rätsch, Gunnar","last_name":"Rätsch"}],"publication_status":"published","citation":{"short":"H. Yèche, G. Dresdner, F. Locatello, M. Hüser, G. Rätsch, in:, Proceedings of 38th International Conference on Machine Learning, ML Research Press, 2021, pp. 11964–11974.","ista":"Yèche H, Dresdner G, Locatello F, Hüser M, Rätsch G. 2021. Neighborhood contrastive learning applied to online patient monitoring. Proceedings of 38th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 139, 11964–11974.","ama":"Yèche H, Dresdner G, Locatello F, Hüser M, Rätsch G. Neighborhood contrastive learning applied to online patient monitoring. In: <i>Proceedings of 38th International Conference on Machine Learning</i>. Vol 139. ML Research Press; 2021:11964-11974.","mla":"Yèche, Hugo, et al. “Neighborhood Contrastive Learning Applied to Online Patient Monitoring.” <i>Proceedings of 38th International Conference on Machine Learning</i>, vol. 139, ML Research Press, 2021, pp. 11964–74.","chicago":"Yèche, Hugo, Gideon Dresdner, Francesco Locatello, Matthias Hüser, and Gunnar Rätsch. “Neighborhood Contrastive Learning Applied to Online Patient Monitoring.” In <i>Proceedings of 38th International Conference on Machine Learning</i>, 139:11964–74. ML Research Press, 2021.","apa":"Yèche, H., Dresdner, G., Locatello, F., Hüser, M., &#38; Rätsch, G. (2021). Neighborhood contrastive learning applied to online patient monitoring. In <i>Proceedings of 38th International Conference on Machine Learning</i> (Vol. 139, pp. 11964–11974). Virtual: ML Research Press.","ieee":"H. Yèche, G. Dresdner, F. Locatello, M. Hüser, and G. Rätsch, “Neighborhood contrastive learning applied to online patient monitoring,” in <i>Proceedings of 38th International Conference on Machine Learning</i>, Virtual, 2021, vol. 139, pp. 11964–11974."},"_id":"14176","extern":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","quality_controlled":"1","oa_version":"Preprint","arxiv":1,"date_updated":"2023-09-11T10:16:55Z","oa":1,"volume":139,"article_processing_charge":"No","title":"Neighborhood contrastive learning applied to online patient monitoring","external_id":{"arxiv":["2106.05142"]},"year":"2021","alternative_title":["PMLR"],"main_file_link":[{"url":"https://arxiv.org/abs/2106.05142","open_access":"1"}],"intvolume":"       139","status":"public","day":"01","type":"conference","publication":"Proceedings of 38th International Conference on Machine Learning","page":"11964-11974","publisher":"ML Research Press","scopus_import":"1","language":[{"iso":"eng"}],"month":"08","date_published":"2021-08-01T00:00:00Z","conference":{"start_date":"2021-07-18","end_date":"2021-07-24","name":"International Conference on Machine Learning","location":"Virtual"},"date_created":"2023-08-22T14:03:04Z","department":[{"_id":"FrLo"}]},{"day":"01","type":"conference","intvolume":"       139","status":"public","publication":"Proceedings of the 38th International Conference on Machine Learning","page":"10401-10412","month":"08","date_published":"2021-08-01T00:00:00Z","publisher":"ML Research Press","scopus_import":"1","language":[{"iso":"eng"}],"department":[{"_id":"FrLo"}],"conference":{"start_date":"2021-07-18","end_date":"2021-07-24","location":"Virtual","name":"ICML: International Conference on Machine Learning"},"date_created":"2023-08-22T14:03:47Z","publication_status":"published","citation":{"chicago":"Träuble, Frederik, Elliot Creager, Niki Kilbertus, Francesco Locatello, Andrea Dittadi, Anirudh Goyal, Bernhard Schölkopf, and Stefan Bauer. “On Disentangled Representations Learned from Correlated Data.” In <i>Proceedings of the 38th International Conference on Machine Learning</i>, 139:10401–12. ML Research Press, 2021.","apa":"Träuble, F., Creager, E., Kilbertus, N., Locatello, F., Dittadi, A., Goyal, A., … Bauer, S. (2021). On disentangled representations learned from correlated data. In <i>Proceedings of the 38th International Conference on Machine Learning</i> (Vol. 139, pp. 10401–10412). Virtual: ML Research Press.","ieee":"F. Träuble <i>et al.</i>, “On disentangled representations learned from correlated data,” in <i>Proceedings of the 38th International Conference on Machine Learning</i>, Virtual, 2021, vol. 139, pp. 10401–10412.","ista":"Träuble F, Creager E, Kilbertus N, Locatello F, Dittadi A, Goyal A, Schölkopf B, Bauer S. 2021. On disentangled representations learned from correlated data. Proceedings of the 38th International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 139, 10401–10412.","short":"F. Träuble, E. Creager, N. Kilbertus, F. Locatello, A. Dittadi, A. Goyal, B. Schölkopf, S. Bauer, in:, Proceedings of the 38th International Conference on Machine Learning, ML Research Press, 2021, pp. 10401–10412.","mla":"Träuble, Frederik, et al. “On Disentangled Representations Learned from Correlated Data.” <i>Proceedings of the 38th International Conference on Machine Learning</i>, vol. 139, ML Research Press, 2021, pp. 10401–12.","ama":"Träuble F, Creager E, Kilbertus N, et al. On disentangled representations learned from correlated data. In: <i>Proceedings of the 38th International Conference on Machine Learning</i>. Vol 139. ML Research Press; 2021:10401-10412."},"abstract":[{"text":"The focus of disentanglement approaches has been on identifying independent factors of variation in data. However, the causal variables underlying real-world observations are often not statistically independent. In this work, we bridge the gap to real-world scenarios by analyzing the behavior of the most prominent disentanglement approaches on correlated data in a large-scale empirical study (including 4260 models). We show and quantify that systematically induced correlations in the dataset are being learned and reflected in the latent representations, which has implications for downstream applications of disentanglement such as fairness. We also demonstrate how to resolve these latent correlations, either using weak supervision during\r\ntraining or by post-hoc correcting a pre-trained model with a small number of labels.","lang":"eng"}],"author":[{"first_name":"Frederik","full_name":"Träuble, Frederik","last_name":"Träuble"},{"first_name":"Elliot","last_name":"Creager","full_name":"Creager, Elliot"},{"first_name":"Niki","full_name":"Kilbertus, Niki","last_name":"Kilbertus"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco","last_name":"Locatello"},{"first_name":"Andrea","last_name":"Dittadi","full_name":"Dittadi, Andrea"},{"first_name":"Anirudh","full_name":"Goyal, Anirudh","last_name":"Goyal"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"first_name":"Stefan","last_name":"Bauer","full_name":"Bauer, Stefan"}],"arxiv":1,"date_updated":"2023-09-11T10:18:48Z","volume":139,"oa":1,"article_processing_charge":"No","_id":"14177","extern":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Published Version","quality_controlled":"1","year":"2021","title":"On disentangled representations learned from correlated data","external_id":{"arxiv":["2006.07886"]},"alternative_title":["PMLR"],"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2006.07886"}]},{"citation":{"chicago":"Dittadi, Andrea, Frederik Träuble, Francesco Locatello, Manuel Wüthrich, Vaibhav Agrawal, Ole Winther, Stefan Bauer, and Bernhard Schölkopf. “On the Transfer of Disentangled Representations in Realistic Settings.” In <i>The Ninth International Conference on Learning Representations</i>, 2021.","apa":"Dittadi, A., Träuble, F., Locatello, F., Wüthrich, M., Agrawal, V., Winther, O., … Schölkopf, B. (2021). On the transfer of disentangled representations in realistic settings. In <i>The Ninth International Conference on Learning Representations</i>. Virtual.","ieee":"A. Dittadi <i>et al.</i>, “On the transfer of disentangled representations in realistic settings,” in <i>The Ninth International Conference on Learning Representations</i>, Virtual, 2021.","ista":"Dittadi A, Träuble F, Locatello F, Wüthrich M, Agrawal V, Winther O, Bauer S, Schölkopf B. 2021. On the transfer of disentangled representations in realistic settings. The Ninth International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","short":"A. Dittadi, F. Träuble, F. Locatello, M. Wüthrich, V. Agrawal, O. Winther, S. Bauer, B. Schölkopf, in:, The Ninth International Conference on Learning Representations, 2021.","mla":"Dittadi, Andrea, et al. “On the Transfer of Disentangled Representations in Realistic Settings.” <i>The Ninth International Conference on Learning Representations</i>, 2021.","ama":"Dittadi A, Träuble F, Locatello F, et al. On the transfer of disentangled representations in realistic settings. In: <i>The Ninth International Conference on Learning Representations</i>. ; 2021."},"day":"04","publication_status":"published","type":"conference","abstract":[{"lang":"eng","text":"Learning meaningful representations that disentangle the underlying structure of the data generating process is considered to be of key importance in machine learning. While disentangled representations were found to be useful for diverse tasks such as abstract reasoning and fair classification, their scalability and real-world impact remain questionable. We introduce a new high-resolution dataset with 1M simulated images and over 1,800 annotated real-world images of the same setup. In contrast to previous work, this new dataset exhibits correlations, a complex underlying structure, and allows to evaluate transfer to unseen simulated and real-world settings where the encoder i) remains in distribution or ii) is out of distribution. We propose new architectures in order to scale disentangled representation learning to realistic high-resolution settings and conduct a large-scale empirical study of disentangled representations on this dataset. We observe that disentanglement is a good predictor for out-of-distribution (OOD) task performance."}],"author":[{"first_name":"Andrea","last_name":"Dittadi","full_name":"Dittadi, Andrea"},{"full_name":"Träuble, Frederik","last_name":"Träuble","first_name":"Frederik"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","full_name":"Locatello, Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683","first_name":"Francesco"},{"last_name":"Wüthrich","full_name":"Wüthrich, Manuel","first_name":"Manuel"},{"first_name":"Vaibhav","full_name":"Agrawal, Vaibhav","last_name":"Agrawal"},{"first_name":"Ole","full_name":"Winther, Ole","last_name":"Winther"},{"last_name":"Bauer","full_name":"Bauer, Stefan","first_name":"Stefan"},{"full_name":"Schölkopf, Bernhard","last_name":"Schölkopf","first_name":"Bernhard"}],"status":"public","publication":"The Ninth International Conference on Learning Representations","arxiv":1,"article_processing_charge":"No","oa":1,"date_updated":"2023-09-11T10:55:30Z","extern":"1","_id":"14178","oa_version":"Preprint","quality_controlled":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","month":"05","year":"2021","date_published":"2021-05-04T00:00:00Z","external_id":{"arxiv":["2010.14407"]},"title":"On the transfer of disentangled representations in realistic settings","language":[{"iso":"eng"}],"main_file_link":[{"url":"https://arxiv.org/abs/2010.14407","open_access":"1"}],"department":[{"_id":"FrLo"}],"conference":{"start_date":"2021-05-03","end_date":"2021-05-07","location":"Virtual","name":"ICLR: International Conference on Learning Representations"},"date_created":"2023-08-22T14:04:16Z"},{"external_id":{"arxiv":["2106.04619"]},"title":"Self-supervised learning with data augmentations provably isolates content from style","year":"2021","main_file_link":[{"url":"https://arxiv.org/abs/2106.04619","open_access":"1"}],"abstract":[{"lang":"eng","text":"Self-supervised representation learning has shown remarkable success in a number of domains. A common practice is to perform data augmentation via hand-crafted transformations intended to leave the semantics of the data invariant. We seek to understand the empirical success of this approach from a theoretical perspective. We formulate the augmentation process as a latent variable model by postulating a partition of the latent representation into a content component, which is assumed invariant to augmentation, and a style component, which is allowed to change. Unlike prior work on disentanglement and independent component analysis, we allow for both nontrivial statistical and causal dependencies in the latent space. We study the identifiability of the latent representation based on pairs of views of the observations and prove sufficient conditions that allow us to identify the invariant content partition up to an invertible mapping in both generative and discriminative settings. We find numerical simulations with dependent latent variables are consistent with our theory. Lastly, we introduce Causal3DIdent, a dataset of high-dimensional, visually complex images with rich causal dependencies, which we use to study the effect of data augmentations performed in practice."}],"author":[{"first_name":"Julius von","full_name":"Kügelgen, Julius von","last_name":"Kügelgen"},{"first_name":"Yash","full_name":"Sharma, Yash","last_name":"Sharma"},{"last_name":"Gresele","full_name":"Gresele, Luigi","first_name":"Luigi"},{"first_name":"Wieland","last_name":"Brendel","full_name":"Brendel, Wieland"},{"first_name":"Bernhard","full_name":"Schölkopf, Bernhard","last_name":"Schölkopf"},{"last_name":"Besserve","full_name":"Besserve, Michel","first_name":"Michel"},{"first_name":"Francesco","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"}],"publication_status":"published","citation":{"mla":"Kügelgen, Julius von, et al. “Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style.” <i>Advances in Neural Information Processing Systems</i>, vol. 34, 2021, pp. 16451–67.","ama":"Kügelgen J von, Sharma Y, Gresele L, et al. Self-supervised learning with data augmentations provably isolates content from style. In: <i>Advances in Neural Information Processing Systems</i>. Vol 34. ; 2021:16451-16467.","ista":"Kügelgen J von, Sharma Y, Gresele L, Brendel W, Schölkopf B, Besserve M, Locatello F. 2021. Self-supervised learning with data augmentations provably isolates content from style. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 34, 16451–16467.","short":"J. von Kügelgen, Y. Sharma, L. Gresele, W. Brendel, B. Schölkopf, M. Besserve, F. Locatello, in:, Advances in Neural Information Processing Systems, 2021, pp. 16451–16467.","ieee":"J. von Kügelgen <i>et al.</i>, “Self-supervised learning with data augmentations provably isolates content from style,” in <i>Advances in Neural Information Processing Systems</i>, Virtual, 2021, vol. 34, pp. 16451–16467.","apa":"Kügelgen, J. von, Sharma, Y., Gresele, L., Brendel, W., Schölkopf, B., Besserve, M., &#38; Locatello, F. (2021). Self-supervised learning with data augmentations provably isolates content from style. In <i>Advances in Neural Information Processing Systems</i> (Vol. 34, pp. 16451–16467). Virtual.","chicago":"Kügelgen, Julius von, Yash Sharma, Luigi Gresele, Wieland Brendel, Bernhard Schölkopf, Michel Besserve, and Francesco Locatello. “Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style.” In <i>Advances in Neural Information Processing Systems</i>, 34:16451–67, 2021."},"_id":"14179","publication_identifier":{"isbn":["9781713845393"]},"extern":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","quality_controlled":"1","oa_version":"Preprint","arxiv":1,"oa":1,"volume":34,"date_updated":"2023-09-11T10:33:19Z","article_processing_charge":"No","language":[{"iso":"eng"}],"month":"06","date_published":"2021-06-08T00:00:00Z","conference":{"location":"Virtual","name":"NeurIPS: Neural Information Processing Systems","end_date":"2021-12-10","start_date":"2021-12-07"},"date_created":"2023-08-22T14:04:36Z","department":[{"_id":"FrLo"}],"intvolume":"        34","status":"public","day":"08","type":"conference","publication":"Advances in Neural Information Processing Systems","page":"16451-16467"},{"quality_controlled":"1","oa_version":"Preprint","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","extern":"1","publication_identifier":{"isbn":["9781713845393"]},"_id":"14180","article_processing_charge":"No","volume":34,"date_updated":"2023-09-11T11:33:46Z","oa":1,"arxiv":1,"author":[{"last_name":"Rahaman","full_name":"Rahaman, Nasim","first_name":"Nasim"},{"full_name":"Gondal, Muhammad Waleed","last_name":"Gondal","first_name":"Muhammad Waleed"},{"full_name":"Joshi, Shruti","last_name":"Joshi","first_name":"Shruti"},{"last_name":"Gehler","full_name":"Gehler, Peter","first_name":"Peter"},{"first_name":"Yoshua","last_name":"Bengio","full_name":"Bengio, Yoshua"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","full_name":"Locatello, Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"}],"abstract":[{"text":"Modern neural network architectures can leverage large amounts of data to generalize well within the training distribution. However, they are less capable of systematic generalization to data drawn from unseen but related distributions, a feat that is hypothesized to require compositional reasoning and reuse of knowledge. In this work, we present Neural Interpreters, an architecture that factorizes inference in a self-attention network as a system of modules, which we call \\emph{functions}. Inputs to the model are routed through a sequence of functions in a way that is end-to-end learned. The proposed architecture can flexibly compose computation along width and depth, and lends itself well to capacity extension after training. To demonstrate the versatility of Neural Interpreters, we evaluate it in two distinct settings: image classification and visual abstract reasoning on Raven Progressive Matrices. In the former, we show that Neural Interpreters perform on par with the vision transformer using fewer parameters, while being transferrable to a new task in a sample efficient manner. In the latter, we find that Neural Interpreters are competitive with respect to the state-of-the-art in terms of systematic generalization. ","lang":"eng"}],"citation":{"ama":"Rahaman N, Gondal MW, Joshi S, et al. Dynamic inference with neural interpreters. In: <i>Advances in Neural Information Processing Systems</i>. Vol 34. ; 2021:10985-10998.","mla":"Rahaman, Nasim, et al. “Dynamic Inference with Neural Interpreters.” <i>Advances in Neural Information Processing Systems</i>, vol. 34, 2021, pp. 10985–98.","ista":"Rahaman N, Gondal MW, Joshi S, Gehler P, Bengio Y, Locatello F, Schölkopf B. 2021. Dynamic inference with neural interpreters. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 34, 10985–10998.","short":"N. Rahaman, M.W. Gondal, S. Joshi, P. Gehler, Y. Bengio, F. Locatello, B. Schölkopf, in:, Advances in Neural Information Processing Systems, 2021, pp. 10985–10998.","apa":"Rahaman, N., Gondal, M. W., Joshi, S., Gehler, P., Bengio, Y., Locatello, F., &#38; Schölkopf, B. (2021). Dynamic inference with neural interpreters. In <i>Advances in Neural Information Processing Systems</i> (Vol. 34, pp. 10985–10998). Virtual.","ieee":"N. Rahaman <i>et al.</i>, “Dynamic inference with neural interpreters,” in <i>Advances in Neural Information Processing Systems</i>, Virtual, 2021, vol. 34, pp. 10985–10998.","chicago":"Rahaman, Nasim, Muhammad Waleed Gondal, Shruti Joshi, Peter Gehler, Yoshua Bengio, Francesco Locatello, and Bernhard Schölkopf. “Dynamic Inference with Neural Interpreters.” In <i>Advances in Neural Information Processing Systems</i>, 34:10985–98, 2021."},"publication_status":"published","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2110.06399"}],"external_id":{"arxiv":["2110.06399"]},"title":"Dynamic inference with neural interpreters","year":"2021","page":"10985-10998","publication":"Advances in Neural Information Processing Systems","status":"public","intvolume":"        34","type":"conference","day":"12","date_created":"2023-08-22T14:04:55Z","conference":{"location":"Virtual","end_date":"2021-12-10","name":"NeurIPS: Neural Information Processing Systems","start_date":"2021-12-07"},"department":[{"_id":"FrLo"}],"language":[{"iso":"eng"}],"date_published":"2021-10-12T00:00:00Z","month":"10"},{"date_published":"2021-05-19T00:00:00Z","month":"05","language":[{"iso":"eng"}],"publisher":"International Joint Conferences on Artificial Intelligence","department":[{"_id":"FrLo"}],"date_created":"2023-08-22T14:05:14Z","conference":{"start_date":"2021-08-19","end_date":"2021-08-27","name":"IJCAI: International Joint Conference on Artificial Intelligence","location":"Montreal, Canada"},"type":"conference","day":"19","status":"public","page":"2337-2343","publication":"Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence","doi":"10.24963/ijcai.2021/322","year":"2021","title":"Boosting variational inference with locally adaptive step-sizes","external_id":{"arxiv":["2105.09240"]},"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2105.09240","open_access":"1"}],"citation":{"chicago":"Dresdner, Gideon, Saurav Shekhar, Fabian Pedregosa, Francesco Locatello, and Gunnar Rätsch. “Boosting Variational Inference with Locally Adaptive Step-Sizes.” In <i>Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence</i>, 2337–43. International Joint Conferences on Artificial Intelligence, 2021. <a href=\"https://doi.org/10.24963/ijcai.2021/322\">https://doi.org/10.24963/ijcai.2021/322</a>.","apa":"Dresdner, G., Shekhar, S., Pedregosa, F., Locatello, F., &#38; Rätsch, G. (2021). Boosting variational inference with locally adaptive step-sizes. In <i>Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence</i> (pp. 2337–2343). Montreal, Canada: International Joint Conferences on Artificial Intelligence. <a href=\"https://doi.org/10.24963/ijcai.2021/322\">https://doi.org/10.24963/ijcai.2021/322</a>","ieee":"G. Dresdner, S. Shekhar, F. Pedregosa, F. Locatello, and G. Rätsch, “Boosting variational inference with locally adaptive step-sizes,” in <i>Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence</i>, Montreal, Canada, 2021, pp. 2337–2343.","ista":"Dresdner G, Shekhar S, Pedregosa F, Locatello F, Rätsch G. 2021. Boosting variational inference with locally adaptive step-sizes. Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. IJCAI: International Joint Conference on Artificial Intelligence, 2337–2343.","short":"G. Dresdner, S. Shekhar, F. Pedregosa, F. Locatello, G. Rätsch, in:, Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, International Joint Conferences on Artificial Intelligence, 2021, pp. 2337–2343.","ama":"Dresdner G, Shekhar S, Pedregosa F, Locatello F, Rätsch G. Boosting variational inference with locally adaptive step-sizes. In: <i>Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence</i>. International Joint Conferences on Artificial Intelligence; 2021:2337-2343. doi:<a href=\"https://doi.org/10.24963/ijcai.2021/322\">10.24963/ijcai.2021/322</a>","mla":"Dresdner, Gideon, et al. “Boosting Variational Inference with Locally Adaptive Step-Sizes.” <i>Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence</i>, International Joint Conferences on Artificial Intelligence, 2021, pp. 2337–43, doi:<a href=\"https://doi.org/10.24963/ijcai.2021/322\">10.24963/ijcai.2021/322</a>."},"publication_status":"published","author":[{"last_name":"Dresdner","full_name":"Dresdner, Gideon","first_name":"Gideon"},{"first_name":"Saurav","full_name":"Shekhar, Saurav","last_name":"Shekhar"},{"full_name":"Pedregosa, Fabian","last_name":"Pedregosa","first_name":"Fabian"},{"first_name":"Francesco","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Gunnar","last_name":"Rätsch","full_name":"Rätsch, Gunnar"}],"abstract":[{"text":"Variational Inference makes a trade-off between the capacity of the variational family and the tractability of finding an approximate posterior distribution. Instead, Boosting Variational Inference allows practitioners to obtain increasingly good posterior approximations by spending more compute. The main obstacle to widespread adoption of Boosting Variational Inference is the amount of resources necessary to improve over a strong Variational Inference baseline. In our work, we trace this limitation back to the global curvature of the KL-divergence. We characterize how the global curvature impacts time and memory consumption, address the problem with the notion of local curvature, and provide a novel approximate backtracking algorithm for estimating local curvature. We give new theoretical convergence rates for our algorithms and provide experimental validation on synthetic and real-world datasets.","lang":"eng"}],"article_processing_charge":"No","date_updated":"2023-09-11T11:14:30Z","oa":1,"arxiv":1,"oa_version":"Published Version","quality_controlled":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication_identifier":{"eisbn":["9780999241196"]},"extern":"1","_id":"14181"},{"page":"116-128","publication":"35th Conference on Neural Information Processing Systems","type":"conference","day":"02","status":"public","intvolume":"        34","department":[{"_id":"FrLo"}],"date_created":"2023-08-22T14:05:41Z","conference":{"end_date":"2021-12-10","name":"NeurIPS: Neural Information Processing Systems","location":"Virtual","start_date":"2021-12-07"},"date_published":"2021-07-02T00:00:00Z","month":"07","language":[{"iso":"eng"}],"date_updated":"2023-09-11T11:31:59Z","oa":1,"volume":34,"article_processing_charge":"No","arxiv":1,"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","quality_controlled":"1","oa_version":"Preprint","_id":"14182","publication_identifier":{"isbn":["9781713845393"]},"extern":"1","publication_status":"published","citation":{"ama":"Träuble F, Kügelgen J von, Kleindessner M, Locatello F, Schölkopf B, Gehler P. Backward-compatible prediction updates: A probabilistic approach. In: <i>35th Conference on Neural Information Processing Systems</i>. Vol 34. ; 2021:116-128.","mla":"Träuble, Frederik, et al. “Backward-Compatible Prediction Updates: A Probabilistic Approach.” <i>35th Conference on Neural Information Processing Systems</i>, vol. 34, 2021, pp. 116–28.","short":"F. Träuble, J. von Kügelgen, M. Kleindessner, F. Locatello, B. Schölkopf, P. Gehler, in:, 35th Conference on Neural Information Processing Systems, 2021, pp. 116–128.","ista":"Träuble F, Kügelgen J von, Kleindessner M, Locatello F, Schölkopf B, Gehler P. 2021. Backward-compatible prediction updates: A probabilistic approach. 35th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 34, 116–128.","ieee":"F. Träuble, J. von Kügelgen, M. Kleindessner, F. Locatello, B. Schölkopf, and P. Gehler, “Backward-compatible prediction updates: A probabilistic approach,” in <i>35th Conference on Neural Information Processing Systems</i>, Virtual, 2021, vol. 34, pp. 116–128.","apa":"Träuble, F., Kügelgen, J. von, Kleindessner, M., Locatello, F., Schölkopf, B., &#38; Gehler, P. (2021). Backward-compatible prediction updates: A probabilistic approach. In <i>35th Conference on Neural Information Processing Systems</i> (Vol. 34, pp. 116–128). Virtual.","chicago":"Träuble, Frederik, Julius von Kügelgen, Matthäus Kleindessner, Francesco Locatello, Bernhard Schölkopf, and Peter Gehler. “Backward-Compatible Prediction Updates: A Probabilistic Approach.” In <i>35th Conference on Neural Information Processing Systems</i>, 34:116–28, 2021."},"author":[{"first_name":"Frederik","full_name":"Träuble, Frederik","last_name":"Träuble"},{"full_name":"Kügelgen, Julius von","last_name":"Kügelgen","first_name":"Julius von"},{"last_name":"Kleindessner","full_name":"Kleindessner, Matthäus","first_name":"Matthäus"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","last_name":"Locatello","full_name":"Locatello, Francesco","first_name":"Francesco"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"last_name":"Gehler","full_name":"Gehler, Peter","first_name":"Peter"}],"abstract":[{"lang":"eng","text":"When machine learning systems meet real world applications, accuracy is only\r\none of several requirements. In this paper, we assay a complementary\r\nperspective originating from the increasing availability of pre-trained and\r\nregularly improving state-of-the-art models. While new improved models develop\r\nat a fast pace, downstream tasks vary more slowly or stay constant. Assume that\r\nwe have a large unlabelled data set for which we want to maintain accurate\r\npredictions. Whenever a new and presumably better ML models becomes available,\r\nwe encounter two problems: (i) given a limited budget, which data points should\r\nbe re-evaluated using the new model?; and (ii) if the new predictions differ\r\nfrom the current ones, should we update? Problem (i) is about compute cost,\r\nwhich matters for very large data sets and models. Problem (ii) is about\r\nmaintaining consistency of the predictions, which can be highly relevant for\r\ndownstream applications; our demand is to avoid negative flips, i.e., changing\r\ncorrect to incorrect predictions. In this paper, we formalize the Prediction\r\nUpdate Problem and present an efficient probabilistic approach as answer to the\r\nabove questions. In extensive experiments on standard classification benchmark\r\ndata sets, we show that our method outperforms alternative strategies along key\r\nmetrics for backward-compatible prediction updates."}],"main_file_link":[{"url":"https://arxiv.org/abs/2107.01057","open_access":"1"}],"year":"2021","external_id":{"arxiv":["2107.01057"]},"title":"Backward-compatible prediction updates: A probabilistic approach"},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","_id":"14221","extern":"1","date_updated":"2023-09-12T07:04:44Z","oa":1,"article_processing_charge":"No","arxiv":1,"publication":"arXiv","author":[{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","last_name":"Locatello","full_name":"Locatello, Francesco","first_name":"Francesco"}],"status":"public","abstract":[{"text":"The world is structured in countless ways. It may be prudent to enforce corresponding structural properties to a learning algorithm's solution, such as incorporating prior beliefs, natural constraints, or causal structures. Doing so may translate to faster, more accurate, and more flexible models, which may directly relate to real-world impact. In this dissertation, we consider two different research areas that concern structuring a learning algorithm's solution: when the structure is known and when it has to be discovered.","lang":"eng"}],"type":"preprint","publication_status":"submitted","citation":{"ista":"Locatello F. Enforcing and discovering structure in machine learning. arXiv, 2111.13693.","short":"F. Locatello, ArXiv (n.d.).","mla":"Locatello, Francesco. “Enforcing and Discovering Structure in Machine Learning.” <i>ArXiv</i>, 2111.13693, doi:<a href=\"https://doi.org/10.48550/arXiv.2111.13693\">10.48550/arXiv.2111.13693</a>.","ama":"Locatello F. Enforcing and discovering structure in machine learning. <i>arXiv</i>. doi:<a href=\"https://doi.org/10.48550/arXiv.2111.13693\">10.48550/arXiv.2111.13693</a>","chicago":"Locatello, Francesco. “Enforcing and Discovering Structure in Machine Learning.” <i>ArXiv</i>, n.d. <a href=\"https://doi.org/10.48550/arXiv.2111.13693\">https://doi.org/10.48550/arXiv.2111.13693</a>.","apa":"Locatello, F. (n.d.). Enforcing and discovering structure in machine learning. <i>arXiv</i>. <a href=\"https://doi.org/10.48550/arXiv.2111.13693\">https://doi.org/10.48550/arXiv.2111.13693</a>","ieee":"F. Locatello, “Enforcing and discovering structure in machine learning,” <i>arXiv</i>. ."},"day":"26","date_created":"2023-08-22T14:23:35Z","department":[{"_id":"FrLo"}],"main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2111.13693"}],"article_number":"2111.13693","language":[{"iso":"eng"}],"title":"Enforcing and discovering structure in machine learning","external_id":{"arxiv":["2111.13693"]},"date_published":"2021-11-26T00:00:00Z","doi":"10.48550/arXiv.2111.13693","year":"2021","month":"11"},{"author":[{"last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Dirk","full_name":"Weissenborn, Dirk","last_name":"Weissenborn"},{"last_name":"Unterthiner","full_name":"Unterthiner, Thomas","first_name":"Thomas"},{"full_name":"Mahendran, Aravindh","last_name":"Mahendran","first_name":"Aravindh"},{"first_name":"Georg","last_name":"Heigold","full_name":"Heigold, Georg"},{"first_name":"Jakob","full_name":"Uszkoreit, Jakob","last_name":"Uszkoreit"},{"first_name":"Alexey","last_name":"Dosovitskiy","full_name":"Dosovitskiy, Alexey"},{"last_name":"Kipf","full_name":"Kipf, Thomas","first_name":"Thomas"}],"status":"public","intvolume":"        33","abstract":[{"text":"Learning object-centric representations of complex scenes is a promising step towards enabling efficient abstract reasoning from low-level perceptual features. Yet, most deep learning approaches learn distributed representations that do not capture the compositional properties of natural scenes. In this paper, we present the Slot Attention module, an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which we call slots. These slots are exchangeable and can bind to any object in the input by specializing through a competitive procedure over multiple rounds of attention. We empirically demonstrate that Slot Attention can extract object-centric representations that enable generalization to unseen compositions when trained on unsupervised object discovery and supervised property prediction tasks.\r\n\r\n","lang":"eng"}],"type":"conference","publication_status":"published","citation":{"mla":"Locatello, Francesco, et al. “Object-Centric Learning with Slot Attention.” <i>Advances in Neural Information Processing Systems</i>, vol. 33, Curran Associates, 2020, pp. 11525–38.","ama":"Locatello F, Weissenborn D, Unterthiner T, et al. Object-centric learning with slot attention. In: <i>Advances in Neural Information Processing Systems</i>. Vol 33. Curran Associates; 2020:11525-11538.","ista":"Locatello F, Weissenborn D, Unterthiner T, Mahendran A, Heigold G, Uszkoreit J, Dosovitskiy A, Kipf T. 2020. Object-centric learning with slot attention. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 33, 11525–11538.","short":"F. Locatello, D. Weissenborn, T. Unterthiner, A. Mahendran, G. Heigold, J. Uszkoreit, A. Dosovitskiy, T. Kipf, in:, Advances in Neural Information Processing Systems, Curran Associates, 2020, pp. 11525–11538.","ieee":"F. Locatello <i>et al.</i>, “Object-centric learning with slot attention,” in <i>Advances in Neural Information Processing Systems</i>, Virtual, 2020, vol. 33, pp. 11525–11538.","apa":"Locatello, F., Weissenborn, D., Unterthiner, T., Mahendran, A., Heigold, G., Uszkoreit, J., … Kipf, T. (2020). Object-centric learning with slot attention. In <i>Advances in Neural Information Processing Systems</i> (Vol. 33, pp. 11525–11538). Virtual: Curran Associates.","chicago":"Locatello, Francesco, Dirk Weissenborn, Thomas Unterthiner, Aravindh Mahendran, Georg Heigold, Jakob Uszkoreit, Alexey Dosovitskiy, and Thomas Kipf. “Object-Centric Learning with Slot Attention.” In <i>Advances in Neural Information Processing Systems</i>, 33:11525–38. Curran Associates, 2020."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","quality_controlled":"1","_id":"14326","extern":"1","publication_identifier":{"isbn":["9781713829546"]},"oa":1,"volume":33,"date_updated":"2023-09-13T12:19:19Z","article_processing_charge":"No","page":"11525-11538","arxiv":1,"publication":"Advances in Neural Information Processing Systems","language":[{"iso":"eng"}],"publisher":"Curran Associates","external_id":{"arxiv":["2006.15055"]},"title":"Object-centric learning with slot attention","date_published":"2020-01-01T00:00:00Z","year":"2020","date_created":"2023-09-13T12:03:46Z","conference":{"start_date":"2020-12-06","end_date":"2020-12-12","location":"Virtual","name":"NeurIPS: Neural Information Processing Systems"},"department":[{"_id":"FrLo"}],"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2006.15055","open_access":"1"}]},{"publication":"Bioinformatics","issue":"Supplement_2","page":"i919-i927","day":"01","type":"journal_article","intvolume":"        36","status":"public","department":[{"_id":"FrLo"}],"date_created":"2023-08-21T12:28:20Z","month":"12","date_published":"2020-12-01T00:00:00Z","article_type":"original","scopus_import":"1","publisher":"Oxford University Press","language":[{"iso":"eng"}],"article_processing_charge":"No","volume":36,"date_updated":"2023-09-11T10:21:00Z","oa":1,"extern":"1","publication_identifier":{"eissn":["1367-4811"]},"_id":"14125","pmid":1,"oa_version":"Published Version","quality_controlled":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","citation":{"short":"S.G. Stark, J. Ficek, F. Locatello, X. Bonilla, S. Chevrier, F. Singer, R. Aebersold, F.S. Al-Quaddoomi, J. Albinus, I. Alborelli, S. Andani, P.-O. Attinger, M. Bacac, D. Baumhoer, B. Beck-Schimmer, N. Beerenwinkel, C. Beisel, L. Bernasconi, A. Bertolini, B. Bodenmiller, X. Bonilla, R. Casanova, S. Chevrier, N. Chicherova, M. D’Costa, E. Danenberg, N. Davidson, M.-A.D. gan, R. Dummer, S. Engler, M. Erkens, K. Eschbach, C. Esposito, A. Fedier, P. Ferreira, J. Ficek, A.L. Frei, B. Frey, S. Goetze, L. Grob, G. Gut, D. Günther, M. Haberecker, P. Haeuptle, V. Heinzelmann-Schwarz, S. Herter, R. Holtackers, T. Huesser, A. Irmisch, F. Jacob, A. Jacobs, T.M. Jaeger, K. Jahn, A.R. James, P.M. Jermann, A. Kahles, A. Kahraman, V.H. Koelzer, W. Kuebler, J. Kuipers, C.P. Kunze, C. Kurzeder, K.-V. Lehmann, M. Levesque, S. Lugert, G. Maass, M. Manz, P. Markolin, J. Mena, U. Menzel, J.M. Metzler, N. Miglino, E.S. Milani, H. Moch, S. Muenst, R. Murri, C.K. Ng, S. Nicolet, M. Nowak, P.G. Pedrioli, L. Pelkmans, S. Piscuoglio, M. Prummer, M. Ritter, C. Rommel, M.L. Rosano-González, G. Rätsch, N. Santacroce, J.S. del Castillo, R. Schlenker, P.C. Schwalie, S. Schwan, T. Schär, G. Senti, F. Singer, S. Sivapatham, B. Snijder, B. Sobottka, V.T. Sreedharan, S. Stark, D.J. Stekhoven, A.P. Theocharides, T.M. Thomas, M. Tolnay, V. Tosevski, N.C. Toussaint, M.A. Tuncel, M. Tusup, A.V. Drogen, M. Vetter, T. Vlajnic, S. Weber, W.P. Weber, R. Wegmann, M. Weller, F. Wendt, N. Wey, A. Wicki, B. Wollscheid, S. Yu, J. Ziegler, M. Zimmermann, M. Zoche, G. Zuend, G. Rätsch, K.-V. Lehmann, Bioinformatics 36 (2020) i919–i927.","ista":"Stark SG et al. 2020. SCIM: Universal single-cell matching with unpaired feature sets. Bioinformatics. 36(Supplement_2), i919–i927.","mla":"Stark, Stefan G., et al. “SCIM: Universal Single-Cell Matching with Unpaired Feature Sets.” <i>Bioinformatics</i>, vol. 36, no. Supplement_2, Oxford University Press, 2020, pp. i919–27, doi:<a href=\"https://doi.org/10.1093/bioinformatics/btaa843\">10.1093/bioinformatics/btaa843</a>.","ama":"Stark SG, Ficek J, Locatello F, et al. SCIM: Universal single-cell matching with unpaired feature sets. <i>Bioinformatics</i>. 2020;36(Supplement_2):i919-i927. doi:<a href=\"https://doi.org/10.1093/bioinformatics/btaa843\">10.1093/bioinformatics/btaa843</a>","chicago":"Stark, Stefan G, Joanna Ficek, Francesco Locatello, Ximena Bonilla, Stéphane Chevrier, Franziska Singer, Rudolf Aebersold, et al. “SCIM: Universal Single-Cell Matching with Unpaired Feature Sets.” <i>Bioinformatics</i>. Oxford University Press, 2020. <a href=\"https://doi.org/10.1093/bioinformatics/btaa843\">https://doi.org/10.1093/bioinformatics/btaa843</a>.","apa":"Stark, S. G., Ficek, J., Locatello, F., Bonilla, X., Chevrier, S., Singer, F., … Lehmann, K.-V. (2020). SCIM: Universal single-cell matching with unpaired feature sets. <i>Bioinformatics</i>. Oxford University Press. <a href=\"https://doi.org/10.1093/bioinformatics/btaa843\">https://doi.org/10.1093/bioinformatics/btaa843</a>","ieee":"S. G. Stark <i>et al.</i>, “SCIM: Universal single-cell matching with unpaired feature sets,” <i>Bioinformatics</i>, vol. 36, no. Supplement_2. Oxford University Press, pp. i919–i927, 2020."},"publication_status":"published","abstract":[{"lang":"eng","text":"Motivation: Recent technological advances have led to an increase in the production and availability of single-cell data. The ability to integrate a set of multi-technology measurements would allow the identification of biologically or clinically meaningful observations through the unification of the perspectives afforded by each technology. In most cases, however, profiling technologies consume the used cells and thus pairwise correspondences between datasets are lost. Due to the sheer size single-cell datasets can acquire, scalable algorithms that are able to universally match single-cell measurements carried out in one cell to its corresponding sibling in another technology are needed.\r\nResults: We propose Single-Cell data Integration via Matching (SCIM), a scalable approach to recover such correspondences in two or more technologies. SCIM assumes that cells share a common (low-dimensional) underlying structure and that the underlying cell distribution is approximately constant across technologies. It constructs a technology-invariant latent space using an autoencoder framework with an adversarial objective. Multi-modal datasets are integrated by pairing cells across technologies using a bipartite matching scheme that operates on the low-dimensional latent representations. We evaluate SCIM on a simulated cellular branching process and show that the cell-to-cell matches derived by SCIM reflect the same pseudotime on the simulated dataset. Moreover, we apply our method to two real-world scenarios, a melanoma tumor sample and a human bone marrow sample, where we pair cells from a scRNA dataset to their sibling cells in a CyTOF dataset achieving 90% and 78% cell-matching accuracy for each one of the samples, respectively."}],"keyword":["Computational Mathematics","Computational Theory and Mathematics","Computer Science Applications","Molecular Biology","Biochemistry","Statistics and Probability"],"author":[{"first_name":"Stefan G","last_name":"Stark","full_name":"Stark, Stefan G"},{"first_name":"Joanna","last_name":"Ficek","full_name":"Ficek, Joanna"},{"full_name":"Locatello, Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Ximena","full_name":"Bonilla, Ximena","last_name":"Bonilla"},{"full_name":"Chevrier, Stéphane","last_name":"Chevrier","first_name":"Stéphane"},{"last_name":"Singer","full_name":"Singer, Franziska","first_name":"Franziska"},{"last_name":"Aebersold","full_name":"Aebersold, Rudolf","first_name":"Rudolf"},{"last_name":"Al-Quaddoomi","full_name":"Al-Quaddoomi, Faisal S","first_name":"Faisal S"},{"first_name":"Jonas","last_name":"Albinus","full_name":"Albinus, Jonas"},{"first_name":"Ilaria","full_name":"Alborelli, Ilaria","last_name":"Alborelli"},{"full_name":"Andani, Sonali","last_name":"Andani","first_name":"Sonali"},{"last_name":"Attinger","full_name":"Attinger, Per-Olof","first_name":"Per-Olof"},{"first_name":"Marina","full_name":"Bacac, Marina","last_name":"Bacac"},{"last_name":"Baumhoer","full_name":"Baumhoer, Daniel","first_name":"Daniel"},{"last_name":"Beck-Schimmer","full_name":"Beck-Schimmer, Beatrice","first_name":"Beatrice"},{"last_name":"Beerenwinkel","full_name":"Beerenwinkel, Niko","first_name":"Niko"},{"first_name":"Christian","full_name":"Beisel, Christian","last_name":"Beisel"},{"last_name":"Bernasconi","full_name":"Bernasconi, Lara","first_name":"Lara"},{"last_name":"Bertolini","full_name":"Bertolini, Anne","first_name":"Anne"},{"first_name":"Bernd","full_name":"Bodenmiller, Bernd","last_name":"Bodenmiller"},{"first_name":"Ximena","last_name":"Bonilla","full_name":"Bonilla, Ximena"},{"full_name":"Casanova, Ruben","last_name":"Casanova","first_name":"Ruben"},{"last_name":"Chevrier","full_name":"Chevrier, Stéphane","first_name":"Stéphane"},{"first_name":"Natalia","last_name":"Chicherova","full_name":"Chicherova, Natalia"},{"first_name":"Maya","last_name":"D'Costa","full_name":"D'Costa, Maya"},{"last_name":"Danenberg","full_name":"Danenberg, Esther","first_name":"Esther"},{"last_name":"Davidson","full_name":"Davidson, Natalie","first_name":"Natalie"},{"first_name":"Monica-Andreea Dră","last_name":"gan","full_name":"gan, Monica-Andreea Dră"},{"last_name":"Dummer","full_name":"Dummer, Reinhard","first_name":"Reinhard"},{"full_name":"Engler, Stefanie","last_name":"Engler","first_name":"Stefanie"},{"first_name":"Martin","last_name":"Erkens","full_name":"Erkens, Martin"},{"first_name":"Katja","last_name":"Eschbach","full_name":"Eschbach, Katja"},{"full_name":"Esposito, Cinzia","last_name":"Esposito","first_name":"Cinzia"},{"first_name":"André","last_name":"Fedier","full_name":"Fedier, André"},{"first_name":"Pedro","last_name":"Ferreira","full_name":"Ferreira, Pedro"},{"first_name":"Joanna","full_name":"Ficek, Joanna","last_name":"Ficek"},{"full_name":"Frei, Anja L","last_name":"Frei","first_name":"Anja L"},{"first_name":"Bruno","full_name":"Frey, Bruno","last_name":"Frey"},{"full_name":"Goetze, Sandra","last_name":"Goetze","first_name":"Sandra"},{"first_name":"Linda","full_name":"Grob, Linda","last_name":"Grob"},{"last_name":"Gut","full_name":"Gut, Gabriele","first_name":"Gabriele"},{"first_name":"Detlef","last_name":"Günther","full_name":"Günther, Detlef"},{"first_name":"Martina","last_name":"Haberecker","full_name":"Haberecker, Martina"},{"last_name":"Haeuptle","full_name":"Haeuptle, Pirmin","first_name":"Pirmin"},{"first_name":"Viola","last_name":"Heinzelmann-Schwarz","full_name":"Heinzelmann-Schwarz, Viola"},{"first_name":"Sylvia","last_name":"Herter","full_name":"Herter, Sylvia"},{"first_name":"Rene","full_name":"Holtackers, Rene","last_name":"Holtackers"},{"first_name":"Tamara","last_name":"Huesser","full_name":"Huesser, Tamara"},{"first_name":"Anja","full_name":"Irmisch, Anja","last_name":"Irmisch"},{"first_name":"Francis","last_name":"Jacob","full_name":"Jacob, Francis"},{"first_name":"Andrea","full_name":"Jacobs, Andrea","last_name":"Jacobs"},{"first_name":"Tim M","last_name":"Jaeger","full_name":"Jaeger, Tim M"},{"last_name":"Jahn","full_name":"Jahn, Katharina","first_name":"Katharina"},{"last_name":"James","full_name":"James, Alva R","first_name":"Alva R"},{"first_name":"Philip M","last_name":"Jermann","full_name":"Jermann, Philip M"},{"full_name":"Kahles, André","last_name":"Kahles","first_name":"André"},{"first_name":"Abdullah","full_name":"Kahraman, Abdullah","last_name":"Kahraman"},{"full_name":"Koelzer, Viktor H","last_name":"Koelzer","first_name":"Viktor H"},{"full_name":"Kuebler, Werner","last_name":"Kuebler","first_name":"Werner"},{"first_name":"Jack","full_name":"Kuipers, Jack","last_name":"Kuipers"},{"full_name":"Kunze, Christian P","last_name":"Kunze","first_name":"Christian P"},{"first_name":"Christian","full_name":"Kurzeder, Christian","last_name":"Kurzeder"},{"full_name":"Lehmann, Kjong-Van","last_name":"Lehmann","first_name":"Kjong-Van"},{"last_name":"Levesque","full_name":"Levesque, Mitchell","first_name":"Mitchell"},{"full_name":"Lugert, Sebastian","last_name":"Lugert","first_name":"Sebastian"},{"last_name":"Maass","full_name":"Maass, Gerd","first_name":"Gerd"},{"full_name":"Manz, Markus","last_name":"Manz","first_name":"Markus"},{"first_name":"Philipp","last_name":"Markolin","full_name":"Markolin, Philipp"},{"full_name":"Mena, Julien","last_name":"Mena","first_name":"Julien"},{"last_name":"Menzel","full_name":"Menzel, Ulrike","first_name":"Ulrike"},{"first_name":"Julian M","full_name":"Metzler, Julian M","last_name":"Metzler"},{"full_name":"Miglino, Nicola","last_name":"Miglino","first_name":"Nicola"},{"first_name":"Emanuela S","full_name":"Milani, Emanuela S","last_name":"Milani"},{"last_name":"Moch","full_name":"Moch, Holger","first_name":"Holger"},{"first_name":"Simone","full_name":"Muenst, Simone","last_name":"Muenst"},{"first_name":"Riccardo","full_name":"Murri, Riccardo","last_name":"Murri"},{"last_name":"Ng","full_name":"Ng, Charlotte KY","first_name":"Charlotte KY"},{"first_name":"Stefan","full_name":"Nicolet, Stefan","last_name":"Nicolet"},{"first_name":"Marta","last_name":"Nowak","full_name":"Nowak, Marta"},{"full_name":"Pedrioli, Patrick GA","last_name":"Pedrioli","first_name":"Patrick GA"},{"first_name":"Lucas","last_name":"Pelkmans","full_name":"Pelkmans, Lucas"},{"last_name":"Piscuoglio","full_name":"Piscuoglio, Salvatore","first_name":"Salvatore"},{"last_name":"Prummer","full_name":"Prummer, Michael","first_name":"Michael"},{"first_name":"Mathilde","full_name":"Ritter, Mathilde","last_name":"Ritter"},{"first_name":"Christian","full_name":"Rommel, Christian","last_name":"Rommel"},{"full_name":"Rosano-González, María L","last_name":"Rosano-González","first_name":"María L"},{"first_name":"Gunnar","full_name":"Rätsch, Gunnar","last_name":"Rätsch"},{"full_name":"Santacroce, Natascha","last_name":"Santacroce","first_name":"Natascha"},{"first_name":"Jacobo Sarabia del","full_name":"Castillo, Jacobo Sarabia del","last_name":"Castillo"},{"first_name":"Ramona","full_name":"Schlenker, Ramona","last_name":"Schlenker"},{"full_name":"Schwalie, Petra C","last_name":"Schwalie","first_name":"Petra C"},{"last_name":"Schwan","full_name":"Schwan, Severin","first_name":"Severin"},{"first_name":"Tobias","last_name":"Schär","full_name":"Schär, Tobias"},{"first_name":"Gabriela","full_name":"Senti, Gabriela","last_name":"Senti"},{"full_name":"Singer, Franziska","last_name":"Singer","first_name":"Franziska"},{"first_name":"Sujana","full_name":"Sivapatham, Sujana","last_name":"Sivapatham"},{"full_name":"Snijder, Berend","last_name":"Snijder","first_name":"Berend"},{"first_name":"Bettina","full_name":"Sobottka, Bettina","last_name":"Sobottka"},{"full_name":"Sreedharan, Vipin T","last_name":"Sreedharan","first_name":"Vipin T"},{"first_name":"Stefan","last_name":"Stark","full_name":"Stark, Stefan"},{"full_name":"Stekhoven, Daniel J","last_name":"Stekhoven","first_name":"Daniel J"},{"full_name":"Theocharides, Alexandre PA","last_name":"Theocharides","first_name":"Alexandre PA"},{"full_name":"Thomas, Tinu M","last_name":"Thomas","first_name":"Tinu M"},{"last_name":"Tolnay","full_name":"Tolnay, Markus","first_name":"Markus"},{"first_name":"Vinko","last_name":"Tosevski","full_name":"Tosevski, Vinko"},{"full_name":"Toussaint, Nora C","last_name":"Toussaint","first_name":"Nora C"},{"first_name":"Mustafa A","full_name":"Tuncel, Mustafa A","last_name":"Tuncel"},{"first_name":"Marina","full_name":"Tusup, Marina","last_name":"Tusup"},{"full_name":"Drogen, Audrey Van","last_name":"Drogen","first_name":"Audrey Van"},{"last_name":"Vetter","full_name":"Vetter, Marcus","first_name":"Marcus"},{"last_name":"Vlajnic","full_name":"Vlajnic, Tatjana","first_name":"Tatjana"},{"full_name":"Weber, Sandra","last_name":"Weber","first_name":"Sandra"},{"full_name":"Weber, Walter P","last_name":"Weber","first_name":"Walter P"},{"first_name":"Rebekka","full_name":"Wegmann, Rebekka","last_name":"Wegmann"},{"full_name":"Weller, Michael","last_name":"Weller","first_name":"Michael"},{"full_name":"Wendt, Fabian","last_name":"Wendt","first_name":"Fabian"},{"first_name":"Norbert","last_name":"Wey","full_name":"Wey, Norbert"},{"full_name":"Wicki, Andreas","last_name":"Wicki","first_name":"Andreas"},{"full_name":"Wollscheid, Bernd","last_name":"Wollscheid","first_name":"Bernd"},{"first_name":"Shuqing","full_name":"Yu, Shuqing","last_name":"Yu"},{"last_name":"Ziegler","full_name":"Ziegler, Johanna","first_name":"Johanna"},{"last_name":"Zimmermann","full_name":"Zimmermann, Marc","first_name":"Marc"},{"first_name":"Martin","last_name":"Zoche","full_name":"Zoche, Martin"},{"first_name":"Gregor","full_name":"Zuend, Gregor","last_name":"Zuend"},{"first_name":"Gunnar","last_name":"Rätsch","full_name":"Rätsch, Gunnar"},{"last_name":"Lehmann","full_name":"Lehmann, Kjong-Van","first_name":"Kjong-Van"}],"main_file_link":[{"url":"https://doi.org/10.1093/bioinformatics/btaa843","open_access":"1"}],"related_material":{"link":[{"relation":"software","url":"https://github.com/ratschlab/scim"}]},"doi":"10.1093/bioinformatics/btaa843","year":"2020","title":"SCIM: Universal single-cell matching with unpaired feature sets","external_id":{"pmid":["33381818"]}},{"intvolume":"        34","status":"public","day":"28","type":"conference","publication":"The 34th AAAI Conference on Artificial Intelligence","issue":"9","page":"13681-13684","scopus_import":"1","publisher":"Association for the Advancement of Artificial Intelligence","language":[{"iso":"eng"}],"month":"07","date_published":"2020-07-28T00:00:00Z","conference":{"start_date":"2020-02-07","end_date":"2020-02-12","location":"New York, NY, United States","name":"AAAI: Conference on Artificial Intelligence"},"date_created":"2023-08-22T14:07:26Z","department":[{"_id":"FrLo"}],"abstract":[{"lang":"eng","text":"The goal of the unsupervised learning of disentangled representations is to\r\nseparate the independent explanatory factors of variation in the data without\r\naccess to supervision. In this paper, we summarize the results of Locatello et\r\nal., 2019, and focus on their implications for practitioners. We discuss the\r\ntheoretical result showing that the unsupervised learning of disentangled\r\nrepresentations is fundamentally impossible without inductive biases and the\r\npractical challenges it entails. Finally, we comment on our experimental\r\nfindings, highlighting the limitations of state-of-the-art approaches and\r\ndirections for future research."}],"author":[{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683","first_name":"Francesco"},{"first_name":"Stefan","full_name":"Bauer, Stefan","last_name":"Bauer"},{"full_name":"Lucic, Mario","last_name":"Lucic","first_name":"Mario"},{"full_name":"Rätsch, Gunnar","last_name":"Rätsch","first_name":"Gunnar"},{"full_name":"Gelly, Sylvain","last_name":"Gelly","first_name":"Sylvain"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"first_name":"Olivier","full_name":"Bachem, Olivier","last_name":"Bachem"}],"citation":{"apa":"Locatello, F., Bauer, S., Lucic, M., Rätsch, G., Gelly, S., Schölkopf, B., &#38; Bachem, O. (2020). A commentary on the unsupervised learning of disentangled representations. In <i>The 34th AAAI Conference on Artificial Intelligence</i> (Vol. 34, pp. 13681–13684). New York, NY, United States: Association for the Advancement of Artificial Intelligence. <a href=\"https://doi.org/10.1609/aaai.v34i09.7120\">https://doi.org/10.1609/aaai.v34i09.7120</a>","ieee":"F. Locatello <i>et al.</i>, “A commentary on the unsupervised learning of disentangled representations,” in <i>The 34th AAAI Conference on Artificial Intelligence</i>, New York, NY, United States, 2020, vol. 34, no. 9, pp. 13681–13684.","chicago":"Locatello, Francesco, Stefan Bauer, Mario Lucic, Gunnar Rätsch, Sylvain Gelly, Bernhard Schölkopf, and Olivier Bachem. “A Commentary on the Unsupervised Learning of Disentangled Representations.” In <i>The 34th AAAI Conference on Artificial Intelligence</i>, 34:13681–84. Association for the Advancement of Artificial Intelligence, 2020. <a href=\"https://doi.org/10.1609/aaai.v34i09.7120\">https://doi.org/10.1609/aaai.v34i09.7120</a>.","ama":"Locatello F, Bauer S, Lucic M, et al. A commentary on the unsupervised learning of disentangled representations. In: <i>The 34th AAAI Conference on Artificial Intelligence</i>. Vol 34. Association for the Advancement of Artificial Intelligence; 2020:13681-13684. doi:<a href=\"https://doi.org/10.1609/aaai.v34i09.7120\">10.1609/aaai.v34i09.7120</a>","mla":"Locatello, Francesco, et al. “A Commentary on the Unsupervised Learning of Disentangled Representations.” <i>The 34th AAAI Conference on Artificial Intelligence</i>, vol. 34, no. 9, Association for the Advancement of Artificial Intelligence, 2020, pp. 13681–84, doi:<a href=\"https://doi.org/10.1609/aaai.v34i09.7120\">10.1609/aaai.v34i09.7120</a>.","ista":"Locatello F, Bauer S, Lucic M, Rätsch G, Gelly S, Schölkopf B, Bachem O. 2020. A commentary on the unsupervised learning of disentangled representations. The 34th AAAI Conference on Artificial Intelligence. AAAI: Conference on Artificial Intelligence vol. 34, 13681–13684.","short":"F. Locatello, S. Bauer, M. Lucic, G. Rätsch, S. Gelly, B. Schölkopf, O. Bachem, in:, The 34th AAAI Conference on Artificial Intelligence, Association for the Advancement of Artificial Intelligence, 2020, pp. 13681–13684."},"publication_status":"published","publication_identifier":{"isbn":["9781577358350"],"eissn":["2374-3468"]},"extern":"1","_id":"14186","oa_version":"Preprint","quality_controlled":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","arxiv":1,"article_processing_charge":"No","date_updated":"2023-09-12T07:44:48Z","oa":1,"volume":34,"external_id":{"arxiv":["2007.14184"]},"title":"A commentary on the unsupervised learning of disentangled representations","year":"2020","doi":"10.1609/aaai.v34i09.7120","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2007.14184"}]},{"main_file_link":[{"url":"https://arxiv.org/abs/2002.11860","open_access":"1"}],"alternative_title":["PMLR"],"year":"2020","external_id":{"arxiv":["2002.11860"]},"title":"Stochastic Frank-Wolfe for constrained finite-sum minimization","article_processing_charge":"No","volume":119,"date_updated":"2023-09-12T08:03:40Z","oa":1,"arxiv":1,"oa_version":"Preprint","quality_controlled":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","extern":"1","_id":"14187","citation":{"short":"G. Négiar, G. Dresdner, A. Tsai, L.E. Ghaoui, F. Locatello, R.M. Freund, F. Pedregosa, in:, Proceedings of the 37th International Conference on Machine Learning, 2020, pp. 7253–7262.","ista":"Négiar G, Dresdner G, Tsai A, Ghaoui LE, Locatello F, Freund RM, Pedregosa F. 2020. Stochastic Frank-Wolfe for constrained finite-sum minimization. Proceedings of the 37th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 119, 7253–7262.","mla":"Négiar, Geoffrey, et al. “Stochastic Frank-Wolfe for Constrained Finite-Sum Minimization.” <i>Proceedings of the 37th International Conference on Machine Learning</i>, vol. 119, 2020, pp. 7253–62.","ama":"Négiar G, Dresdner G, Tsai A, et al. Stochastic Frank-Wolfe for constrained finite-sum minimization. In: <i>Proceedings of the 37th International Conference on Machine Learning</i>. Vol 119. ; 2020:7253-7262.","chicago":"Négiar, Geoffrey, Gideon Dresdner, Alicia Tsai, Laurent El Ghaoui, Francesco Locatello, Robert M. Freund, and Fabian Pedregosa. “Stochastic Frank-Wolfe for Constrained Finite-Sum Minimization.” In <i>Proceedings of the 37th International Conference on Machine Learning</i>, 119:7253–62, 2020.","ieee":"G. Négiar <i>et al.</i>, “Stochastic Frank-Wolfe for constrained finite-sum minimization,” in <i>Proceedings of the 37th International Conference on Machine Learning</i>, Virtual, 2020, vol. 119, pp. 7253–7262.","apa":"Négiar, G., Dresdner, G., Tsai, A., Ghaoui, L. E., Locatello, F., Freund, R. M., &#38; Pedregosa, F. (2020). Stochastic Frank-Wolfe for constrained finite-sum minimization. In <i>Proceedings of the 37th International Conference on Machine Learning</i> (Vol. 119, pp. 7253–7262). Virtual."},"publication_status":"published","author":[{"last_name":"Négiar","full_name":"Négiar, Geoffrey","first_name":"Geoffrey"},{"last_name":"Dresdner","full_name":"Dresdner, Gideon","first_name":"Gideon"},{"last_name":"Tsai","full_name":"Tsai, Alicia","first_name":"Alicia"},{"first_name":"Laurent El","full_name":"Ghaoui, Laurent El","last_name":"Ghaoui"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683"},{"full_name":"Freund, Robert M.","last_name":"Freund","first_name":"Robert M."},{"first_name":"Fabian","last_name":"Pedregosa","full_name":"Pedregosa, Fabian"}],"abstract":[{"lang":"eng","text":"We propose a novel Stochastic Frank-Wolfe (a.k.a. conditional gradient)\r\nalgorithm for constrained smooth finite-sum minimization with a generalized\r\nlinear prediction/structure. This class of problems includes empirical risk\r\nminimization with sparse, low-rank, or other structured constraints. The\r\nproposed method is simple to implement, does not require step-size tuning, and\r\nhas a constant per-iteration cost that is independent of the dataset size.\r\nFurthermore, as a byproduct of the method we obtain a stochastic estimator of\r\nthe Frank-Wolfe gap that can be used as a stopping criterion. Depending on the\r\nsetting, the proposed method matches or improves on the best computational\r\nguarantees for Stochastic Frank-Wolfe algorithms. Benchmarks on several\r\ndatasets highlight different regimes in which the proposed method exhibits a\r\nfaster empirical convergence than related methods. Finally, we provide an\r\nimplementation of all considered methods in an open-source package."}],"department":[{"_id":"FrLo"}],"date_created":"2023-08-22T14:07:52Z","conference":{"name":"International Conference on Machine Learning","location":"Virtual","end_date":"2020-07-18","start_date":"2020-07-13"},"date_published":"2020-07-27T00:00:00Z","month":"07","language":[{"iso":"eng"}],"page":"7253-7262","publication":"Proceedings of the 37th International Conference on Machine Learning","type":"conference","day":"27","status":"public","intvolume":"       119"},{"type":"conference","day":"07","status":"public","intvolume":"       119","page":"6348–6359","publication":"Proceedings of the 37th International Conference on Machine Learning","date_published":"2020-07-07T00:00:00Z","month":"07","language":[{"iso":"eng"}],"scopus_import":"1","department":[{"_id":"FrLo"}],"date_created":"2023-08-22T14:08:14Z","conference":{"start_date":"2020-07-13","location":"Virtual","end_date":"2020-07-18","name":"International Conference on Machine Learning"},"citation":{"chicago":"Locatello, Francesco, Ben Poole, Gunnar Rätsch, Bernhard Schölkopf, Olivier Bachem, and Michael Tschannen. “Weakly-Supervised Disentanglement without Compromises.” In <i>Proceedings of the 37th International Conference on Machine Learning</i>, 119:6348–6359, 2020.","ieee":"F. Locatello, B. Poole, G. Rätsch, B. Schölkopf, O. Bachem, and M. Tschannen, “Weakly-supervised disentanglement without compromises,” in <i>Proceedings of the 37th International Conference on Machine Learning</i>, Virtual, 2020, vol. 119, pp. 6348–6359.","apa":"Locatello, F., Poole, B., Rätsch, G., Schölkopf, B., Bachem, O., &#38; Tschannen, M. (2020). Weakly-supervised disentanglement without compromises. In <i>Proceedings of the 37th International Conference on Machine Learning</i> (Vol. 119, pp. 6348–6359). Virtual.","ista":"Locatello F, Poole B, Rätsch G, Schölkopf B, Bachem O, Tschannen M. 2020. Weakly-supervised disentanglement without compromises. Proceedings of the 37th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 119, 6348–6359.","short":"F. Locatello, B. Poole, G. Rätsch, B. Schölkopf, O. Bachem, M. Tschannen, in:, Proceedings of the 37th International Conference on Machine Learning, 2020, pp. 6348–6359.","ama":"Locatello F, Poole B, Rätsch G, Schölkopf B, Bachem O, Tschannen M. Weakly-supervised disentanglement without compromises. In: <i>Proceedings of the 37th International Conference on Machine Learning</i>. Vol 119. ; 2020:6348–6359.","mla":"Locatello, Francesco, et al. “Weakly-Supervised Disentanglement without Compromises.” <i>Proceedings of the 37th International Conference on Machine Learning</i>, vol. 119, 2020, pp. 6348–6359."},"publication_status":"published","author":[{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683"},{"full_name":"Poole, Ben","last_name":"Poole","first_name":"Ben"},{"last_name":"Rätsch","full_name":"Rätsch, Gunnar","first_name":"Gunnar"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"full_name":"Bachem, Olivier","last_name":"Bachem","first_name":"Olivier"},{"first_name":"Michael","last_name":"Tschannen","full_name":"Tschannen, Michael"}],"abstract":[{"lang":"eng","text":"Intelligent agents should be able to learn useful representations by\r\nobserving changes in their environment. We model such observations as pairs of\r\nnon-i.i.d. images sharing at least one of the underlying factors of variation.\r\nFirst, we theoretically show that only knowing how many factors have changed,\r\nbut not which ones, is sufficient to learn disentangled representations.\r\nSecond, we provide practical algorithms that learn disentangled representations\r\nfrom pairs of images without requiring annotation of groups, individual\r\nfactors, or the number of factors that have changed. Third, we perform a\r\nlarge-scale empirical study and show that such pairs of observations are\r\nsufficient to reliably learn disentangled representations on several benchmark\r\ndata sets. Finally, we evaluate our learned representations and find that they\r\nare simultaneously useful on a diverse suite of tasks, including generalization\r\nunder covariate shifts, fairness, and abstract reasoning. Overall, our results\r\ndemonstrate that weak supervision enables learning of useful disentangled\r\nrepresentations in realistic scenarios."}],"article_processing_charge":"No","date_updated":"2023-09-12T07:59:29Z","oa":1,"volume":119,"arxiv":1,"quality_controlled":"1","oa_version":"Preprint","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","extern":"1","_id":"14188","year":"2020","external_id":{"arxiv":["2002.02886"]},"title":"Weakly-supervised disentanglement without compromises","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2002.02886"}],"alternative_title":["PMLR"]},{"department":[{"_id":"FrLo"}],"license":"https://creativecommons.org/licenses/by/4.0/","has_accepted_license":"1","date_created":"2023-08-22T14:10:34Z","date_published":"2020-09-01T00:00:00Z","article_type":"original","month":"09","language":[{"iso":"eng"}],"publisher":"MIT Press","scopus_import":"1","publication":"Journal of Machine Learning Research","type":"journal_article","day":"01","status":"public","intvolume":"        21","main_file_link":[{"url":"https://jmlr.csail.mit.edu/papers/v21/19-976.html","open_access":"1"}],"article_number":"209","tmp":{"image":"/images/cc_by.png","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)"},"ddc":["000"],"year":"2020","title":"A sober look at the unsupervised learning of disentangled representations and their evaluation","external_id":{"arxiv":["2010.14766"]},"oa":1,"date_updated":"2023-09-12T09:23:56Z","volume":21,"article_processing_charge":"No","arxiv":1,"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Published Version","quality_controlled":"1","_id":"14195","extern":"1","publication_status":"published","citation":{"ieee":"F. Locatello <i>et al.</i>, “A sober look at the unsupervised learning of disentangled representations and their evaluation,” <i>Journal of Machine Learning Research</i>, vol. 21. MIT Press, 2020.","apa":"Locatello, F., Bauer, S., Lucic, M., Rätsch, G., Gelly, S., Schölkopf, B., &#38; Bachem, O. (2020). A sober look at the unsupervised learning of disentangled representations and their evaluation. <i>Journal of Machine Learning Research</i>. MIT Press.","chicago":"Locatello, Francesco, Stefan Bauer, Mario Lucic, Gunnar Rätsch, Sylvain Gelly, Bernhard Schölkopf, and Olivier Bachem. “A Sober Look at the Unsupervised Learning of Disentangled Representations and Their Evaluation.” <i>Journal of Machine Learning Research</i>. MIT Press, 2020.","ama":"Locatello F, Bauer S, Lucic M, et al. A sober look at the unsupervised learning of disentangled representations and their evaluation. <i>Journal of Machine Learning Research</i>. 2020;21.","mla":"Locatello, Francesco, et al. “A Sober Look at the Unsupervised Learning of Disentangled Representations and Their Evaluation.” <i>Journal of Machine Learning Research</i>, vol. 21, 209, MIT Press, 2020.","short":"F. Locatello, S. Bauer, M. Lucic, G. Rätsch, S. Gelly, B. Schölkopf, O. Bachem, Journal of Machine Learning Research 21 (2020).","ista":"Locatello F, Bauer S, Lucic M, Rätsch G, Gelly S, Schölkopf B, Bachem O. 2020. A sober look at the unsupervised learning of disentangled representations and their evaluation. Journal of Machine Learning Research. 21, 209."},"author":[{"first_name":"Francesco","orcid":"0000-0002-4850-0683","last_name":"Locatello","full_name":"Locatello, Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Stefan","last_name":"Bauer","full_name":"Bauer, Stefan"},{"first_name":"Mario","full_name":"Lucic, Mario","last_name":"Lucic"},{"first_name":"Gunnar","last_name":"Rätsch","full_name":"Rätsch, Gunnar"},{"first_name":"Sylvain","last_name":"Gelly","full_name":"Gelly, Sylvain"},{"first_name":"Bernhard","last_name":"Schölkopf","full_name":"Schölkopf, Bernhard"},{"first_name":"Olivier","last_name":"Bachem","full_name":"Bachem, Olivier"}],"abstract":[{"text":"The idea behind the unsupervised learning of disentangled representations is that real-world data is generated by a few explanatory factors of variation which can be recovered by unsupervised learning algorithms. In this paper, we provide a sober look at recent progress in the field and challenge some common assumptions. We first theoretically show that the unsupervised learning of disentangled representations is fundamentally impossible without inductive biases on both the models and the data. Then, we train over 14000\r\n models covering most prominent methods and evaluation metrics in a reproducible large-scale experimental study on eight data sets. We observe that while the different methods successfully enforce properties “encouraged” by the corresponding losses, well-disentangled models seemingly cannot be identified without supervision. Furthermore, different evaluation metrics do not always agree on what should be considered “disentangled” and exhibit systematic differences in the estimation. Finally, increased disentanglement does not seem to necessarily lead to a decreased sample complexity of learning for downstream tasks. Our results suggest that future work on disentanglement learning should be explicit about the role of inductive biases and (implicit) supervision, investigate concrete benefits of enforcing disentanglement of the learned representations, and consider a reproducible experimental setup covering several data sets.","lang":"eng"}]},{"publication_status":"published","citation":{"ieee":"F. Locatello, M. Tschannen, S. Bauer, G. Rätsch, B. Schölkopf, and O. Bachem, “Disentangling factors of variation using few labels,” in <i>8th International Conference on Learning Representations</i>, Virtual, 2019.","apa":"Locatello, F., Tschannen, M., Bauer, S., Rätsch, G., Schölkopf, B., &#38; Bachem, O. (2019). Disentangling factors of variation using few labels. In <i>8th International Conference on Learning Representations</i>. Virtual.","chicago":"Locatello, Francesco, Michael Tschannen, Stefan Bauer, Gunnar Rätsch, Bernhard Schölkopf, and Olivier Bachem. “Disentangling Factors of Variation Using Few Labels.” In <i>8th International Conference on Learning Representations</i>, 2019.","mla":"Locatello, Francesco, et al. “Disentangling Factors of Variation Using Few Labels.” <i>8th International Conference on Learning Representations</i>, 2019.","ama":"Locatello F, Tschannen M, Bauer S, Rätsch G, Schölkopf B, Bachem O. Disentangling factors of variation using few labels. In: <i>8th International Conference on Learning Representations</i>. ; 2019.","short":"F. Locatello, M. Tschannen, S. Bauer, G. Rätsch, B. Schölkopf, O. Bachem, in:, 8th International Conference on Learning Representations, 2019.","ista":"Locatello F, Tschannen M, Bauer S, Rätsch G, Schölkopf B, Bachem O. 2019. Disentangling factors of variation using few labels. 8th International Conference on Learning Representations. ICLR: International Conference on Learning Representations."},"day":"20","type":"conference","abstract":[{"lang":"eng","text":"Learning disentangled representations is considered a cornerstone problem in\r\nrepresentation learning. Recently, Locatello et al. (2019) demonstrated that\r\nunsupervised disentanglement learning without inductive biases is theoretically\r\nimpossible and that existing inductive biases and unsupervised methods do not\r\nallow to consistently learn disentangled representations. However, in many\r\npractical settings, one might have access to a limited amount of supervision,\r\nfor example through manual labeling of (some) factors of variation in a few\r\ntraining examples. In this paper, we investigate the impact of such supervision\r\non state-of-the-art disentanglement methods and perform a large scale study,\r\ntraining over 52000 models under well-defined and reproducible experimental\r\nconditions. We observe that a small number of labeled examples (0.01--0.5\\% of\r\nthe data set), with potentially imprecise and incomplete labels, is sufficient\r\nto perform model selection on state-of-the-art unsupervised models. Further, we\r\ninvestigate the benefit of incorporating supervision into the training process.\r\nOverall, we empirically validate that with little and imprecise supervision it\r\nis possible to reliably learn disentangled representations."}],"status":"public","author":[{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683","first_name":"Francesco"},{"first_name":"Michael","full_name":"Tschannen, Michael","last_name":"Tschannen"},{"last_name":"Bauer","full_name":"Bauer, Stefan","first_name":"Stefan"},{"full_name":"Rätsch, Gunnar","last_name":"Rätsch","first_name":"Gunnar"},{"first_name":"Bernhard","last_name":"Schölkopf","full_name":"Schölkopf, Bernhard"},{"first_name":"Olivier","last_name":"Bachem","full_name":"Bachem, Olivier"}],"arxiv":1,"publication":"8th International Conference on Learning Representations","date_updated":"2023-09-12T07:01:34Z","oa":1,"article_processing_charge":"No","_id":"14184","extern":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","quality_controlled":"1","oa_version":"Preprint","year":"2019","month":"12","date_published":"2019-12-20T00:00:00Z","external_id":{"arxiv":["1905.01258"]},"title":"Disentangling factors of variation using few labels","scopus_import":"1","language":[{"iso":"eng"}],"department":[{"_id":"FrLo"}],"main_file_link":[{"url":"https://arxiv.org/abs/1905.01258","open_access":"1"}],"conference":{"start_date":"2020-04-26","end_date":"2020-05-01","name":"ICLR: International Conference on Learning Representations","location":"Virtual"},"date_created":"2023-08-22T14:06:37Z"},{"_id":"14189","extern":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","quality_controlled":"1","oa_version":"Preprint","arxiv":1,"date_updated":"2023-09-12T08:07:38Z","volume":115,"oa":1,"article_processing_charge":"No","abstract":[{"text":"We consider the problem of recovering a common latent source with independent\r\ncomponents from multiple views. This applies to settings in which a variable is\r\nmeasured with multiple experimental modalities, and where the goal is to\r\nsynthesize the disparate measurements into a single unified representation. We\r\nconsider the case that the observed views are a nonlinear mixing of\r\ncomponent-wise corruptions of the sources. When the views are considered\r\nseparately, this reduces to nonlinear Independent Component Analysis (ICA) for\r\nwhich it is provably impossible to undo the mixing. We present novel\r\nidentifiability proofs that this is possible when the multiple views are\r\nconsidered jointly, showing that the mixing can theoretically be undone using\r\nfunction approximators such as deep neural networks. In contrast to known\r\nidentifiability results for nonlinear ICA, we prove that independent latent\r\nsources with arbitrary mixing can be recovered as long as multiple,\r\nsufficiently different noisy views are available.","lang":"eng"}],"author":[{"first_name":"Luigi","last_name":"Gresele","full_name":"Gresele, Luigi"},{"first_name":"Paul K.","last_name":"Rubenstein","full_name":"Rubenstein, Paul K."},{"full_name":"Mehrjou, Arash","last_name":"Mehrjou","first_name":"Arash"},{"first_name":"Francesco","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Bernhard","full_name":"Schölkopf, Bernhard","last_name":"Schölkopf"}],"publication_status":"published","citation":{"chicago":"Gresele, Luigi, Paul K. Rubenstein, Arash Mehrjou, Francesco Locatello, and Bernhard Schölkopf. “The Incomplete Rosetta Stone Problem: Identifiability Results for Multi-View Nonlinear ICA.” In <i>Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence</i>, 115:217–27. ML Research Press, 2019.","apa":"Gresele, L., Rubenstein, P. K., Mehrjou, A., Locatello, F., &#38; Schölkopf, B. (2019). The incomplete Rosetta Stone problem: Identifiability results for multi-view nonlinear ICA. In <i>Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence</i> (Vol. 115, pp. 217–227). Tel Aviv, Israel: ML Research Press.","ieee":"L. Gresele, P. K. Rubenstein, A. Mehrjou, F. Locatello, and B. Schölkopf, “The incomplete Rosetta Stone problem: Identifiability results for multi-view nonlinear ICA,” in <i>Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence</i>, Tel Aviv, Israel, 2019, vol. 115, pp. 217–227.","ista":"Gresele L, Rubenstein PK, Mehrjou A, Locatello F, Schölkopf B. 2019. The incomplete Rosetta Stone problem: Identifiability results for multi-view nonlinear ICA. Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence. UAI: Uncertainty in Artificial Intelligence, PMLR, vol. 115, 217–227.","short":"L. Gresele, P.K. Rubenstein, A. Mehrjou, F. Locatello, B. Schölkopf, in:, Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence, ML Research Press, 2019, pp. 217–227.","mla":"Gresele, Luigi, et al. “The Incomplete Rosetta Stone Problem: Identifiability Results for Multi-View Nonlinear ICA.” <i>Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence</i>, vol. 115, ML Research Press, 2019, pp. 217–27.","ama":"Gresele L, Rubenstein PK, Mehrjou A, Locatello F, Schölkopf B. The incomplete Rosetta Stone problem: Identifiability results for multi-view nonlinear ICA. In: <i>Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence</i>. Vol 115. ML Research Press; 2019:217-227."},"alternative_title":["PMLR"],"main_file_link":[{"url":"https://arxiv.org/abs/1905.06642","open_access":"1"}],"title":"The incomplete Rosetta Stone problem: Identifiability results for multi-view nonlinear ICA","external_id":{"arxiv":["1905.06642"]},"year":"2019","publication":"Proceedings of the 35th Conference on Uncertainty in Artificial  Intelligence","page":"217-227","intvolume":"       115","status":"public","day":"16","type":"conference","conference":{"start_date":"2019-07-22","end_date":"2019-07-25","name":"UAI: Uncertainty in Artificial Intelligence","location":"Tel Aviv, Israel"},"date_created":"2023-08-22T14:08:35Z","department":[{"_id":"FrLo"}],"publisher":"ML Research Press","scopus_import":"1","language":[{"iso":"eng"}],"month":"05","date_published":"2019-05-16T00:00:00Z"},{"conference":{"location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems","end_date":"2019-12-14","start_date":"2019-12-08"},"date_created":"2023-08-22T14:09:13Z","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1906.03292"}],"department":[{"_id":"FrLo"}],"external_id":{"arxiv":["1906.03292"]},"title":"On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset","language":[{"iso":"eng"}],"month":"06","year":"2019","date_published":"2019-06-07T00:00:00Z","extern":"1","publication_identifier":{"isbn":["9781713807933"]},"_id":"14190","oa_version":"Preprint","quality_controlled":"1","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","publication":"Advances in Neural Information Processing Systems","arxiv":1,"article_processing_charge":"No","volume":32,"oa":1,"date_updated":"2023-09-13T09:46:38Z","abstract":[{"lang":"eng","text":"Learning meaningful and compact representations with disentangled semantic\r\naspects is considered to be of key importance in representation learning. Since\r\nreal-world data is notoriously costly to collect, many recent state-of-the-art\r\ndisentanglement models have heavily relied on synthetic toy data-sets. In this\r\npaper, we propose a novel data-set which consists of over one million images of\r\nphysical 3D objects with seven factors of variation, such as object color,\r\nshape, size and position. In order to be able to control all the factors of\r\nvariation precisely, we built an experimental platform where the objects are\r\nbeing moved by a robotic arm. In addition, we provide two more datasets which\r\nconsist of simulations of the experimental setup. These datasets provide for\r\nthe first time the possibility to systematically investigate how well different\r\ndisentanglement methods perform on real data in comparison to simulation, and\r\nhow simulated data can be leveraged to build better representations of the real\r\nworld. We provide a first experimental study of these questions and our results\r\nindicate that learned models transfer poorly, but that model and hyperparameter\r\nselection is an effective means of transferring information to the real world."}],"intvolume":"        32","author":[{"full_name":"Gondal, Muhammad Waleed","last_name":"Gondal","first_name":"Muhammad Waleed"},{"full_name":"Wüthrich, Manuel","last_name":"Wüthrich","first_name":"Manuel"},{"first_name":"Đorđe","full_name":"Miladinović, Đorđe","last_name":"Miladinović"},{"full_name":"Locatello, Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"full_name":"Breidt, Martin","last_name":"Breidt","first_name":"Martin"},{"last_name":"Volchkov","full_name":"Volchkov, Valentin","first_name":"Valentin"},{"first_name":"Joel","last_name":"Akpo","full_name":"Akpo, Joel"},{"full_name":"Bachem, Olivier","last_name":"Bachem","first_name":"Olivier"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"full_name":"Bauer, Stefan","last_name":"Bauer","first_name":"Stefan"}],"status":"public","citation":{"mla":"Gondal, Muhammad Waleed, et al. “On the Transfer of Inductive Bias from Simulation to the Real World: A New Disentanglement Dataset.” <i>Advances in Neural Information Processing Systems</i>, vol. 32, 2019.","ama":"Gondal MW, Wüthrich M, Miladinović Đ, et al. On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset. In: <i>Advances in Neural Information Processing Systems</i>. Vol 32. ; 2019.","short":"M.W. Gondal, M. Wüthrich, Đ. Miladinović, F. Locatello, M. Breidt, V. Volchkov, J. Akpo, O. Bachem, B. Schölkopf, S. Bauer, in:, Advances in Neural Information Processing Systems, 2019.","ista":"Gondal MW, Wüthrich M, Miladinović Đ, Locatello F, Breidt M, Volchkov V, Akpo J, Bachem O, Schölkopf B, Bauer S. 2019. On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 32.","apa":"Gondal, M. W., Wüthrich, M., Miladinović, Đ., Locatello, F., Breidt, M., Volchkov, V., … Bauer, S. (2019). On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset. In <i>Advances in Neural Information Processing Systems</i> (Vol. 32). Vancouver, Canada.","ieee":"M. W. Gondal <i>et al.</i>, “On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2019, vol. 32.","chicago":"Gondal, Muhammad Waleed, Manuel Wüthrich, Đorđe Miladinović, Francesco Locatello, Martin Breidt, Valentin Volchkov, Joel Akpo, Olivier Bachem, Bernhard Schölkopf, and Stefan Bauer. “On the Transfer of Inductive Bias from Simulation to the Real World: A New Disentanglement Dataset.” In <i>Advances in Neural Information Processing Systems</i>, Vol. 32, 2019."},"day":"07","publication_status":"published","type":"conference"},{"intvolume":"        32","status":"public","day":"29","type":"conference","publication":"Advances in Neural Information Processing Systems","page":"14291–14301","scopus_import":"1","language":[{"iso":"eng"}],"month":"12","date_published":"2019-12-29T00:00:00Z","conference":{"start_date":"2019-12-08","end_date":"2019-12-14","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems"},"date_created":"2023-08-22T14:09:35Z","department":[{"_id":"FrLo"}],"abstract":[{"text":"A broad class of convex optimization problems can be formulated as a semidefinite program (SDP), minimization of a convex function over the positive-semidefinite cone subject to some affine constraints. The majority of classical SDP solvers are designed for the deterministic setting where problem data is readily available. In this setting, generalized conditional gradient methods (aka Frank-Wolfe-type methods) provide scalable solutions by leveraging the so-called linear minimization oracle instead of the projection onto the semidefinite cone. Most problems in machine learning and modern engineering applications, however, contain some degree of stochasticity. In this work, we propose the first conditional-gradient-type method for solving stochastic optimization problems under affine constraints. Our method guarantees O(k−1/3) convergence rate in expectation on the objective residual and O(k−5/12) on the feasibility gap.","lang":"eng"}],"author":[{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","last_name":"Locatello","full_name":"Locatello, Francesco","first_name":"Francesco"},{"last_name":"Yurtsever","full_name":"Yurtsever, Alp","first_name":"Alp"},{"full_name":"Fercoq, Olivier","last_name":"Fercoq","first_name":"Olivier"},{"last_name":"Cevher","full_name":"Cevher, Volkan","first_name":"Volkan"}],"citation":{"ieee":"F. Locatello, A. Yurtsever, O. Fercoq, and V. Cevher, “Stochastic Frank-Wolfe for composite convex minimization,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2019, vol. 32, pp. 14291–14301.","apa":"Locatello, F., Yurtsever, A., Fercoq, O., &#38; Cevher, V. (2019). Stochastic Frank-Wolfe for composite convex minimization. In <i>Advances in Neural Information Processing Systems</i> (Vol. 32, pp. 14291–14301). Vancouver, Canada.","chicago":"Locatello, Francesco, Alp Yurtsever, Olivier Fercoq, and Volkan Cevher. “Stochastic Frank-Wolfe for Composite Convex Minimization.” In <i>Advances in Neural Information Processing Systems</i>, 32:14291–14301, 2019.","ama":"Locatello F, Yurtsever A, Fercoq O, Cevher V. Stochastic Frank-Wolfe for composite convex minimization. In: <i>Advances in Neural Information Processing Systems</i>. Vol 32. ; 2019:14291–14301.","mla":"Locatello, Francesco, et al. “Stochastic Frank-Wolfe for Composite Convex Minimization.” <i>Advances in Neural Information Processing Systems</i>, vol. 32, 2019, pp. 14291–14301.","short":"F. Locatello, A. Yurtsever, O. Fercoq, V. Cevher, in:, Advances in Neural Information Processing Systems, 2019, pp. 14291–14301.","ista":"Locatello F, Yurtsever A, Fercoq O, Cevher V. 2019. Stochastic Frank-Wolfe for composite convex minimization. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 32, 14291–14301."},"publication_status":"published","publication_identifier":{"isbn":["9781713807933"]},"extern":"1","_id":"14191","oa_version":"Preprint","quality_controlled":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","arxiv":1,"article_processing_charge":"No","date_updated":"2023-09-12T08:48:45Z","oa":1,"volume":32,"external_id":{"arxiv":["1901.10348"]},"title":"Stochastic Frank-Wolfe for composite convex minimization","year":"2019","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1901.10348"}]},{"type":"conference","publication_status":"published","citation":{"short":"S. van Steenkiste, F. Locatello, J. Schmidhuber, O. Bachem, in:, Advances in Neural Information Processing Systems, 2019.","ista":"Steenkiste S van, Locatello F, Schmidhuber J, Bachem O. 2019. Are disentangled representations helpful for abstract visual reasoning? Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 32.","mla":"Steenkiste, Sjoerd van, et al. “Are Disentangled Representations Helpful for Abstract Visual Reasoning?” <i>Advances in Neural Information Processing Systems</i>, vol. 32, 2019.","ama":"Steenkiste S van, Locatello F, Schmidhuber J, Bachem O. Are disentangled representations helpful for abstract visual reasoning? In: <i>Advances in Neural Information Processing Systems</i>. Vol 32. ; 2019.","chicago":"Steenkiste, Sjoerd van, Francesco Locatello, Jürgen Schmidhuber, and Olivier Bachem. “Are Disentangled Representations Helpful for Abstract Visual Reasoning?” In <i>Advances in Neural Information Processing Systems</i>, Vol. 32, 2019.","ieee":"S. van Steenkiste, F. Locatello, J. Schmidhuber, and O. Bachem, “Are disentangled representations helpful for abstract visual reasoning?,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2019, vol. 32.","apa":"Steenkiste, S. van, Locatello, F., Schmidhuber, J., &#38; Bachem, O. (2019). Are disentangled representations helpful for abstract visual reasoning? In <i>Advances in Neural Information Processing Systems</i> (Vol. 32). Vancouver, Canada."},"day":"29","author":[{"last_name":"Steenkiste","full_name":"Steenkiste, Sjoerd van","first_name":"Sjoerd van"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683"},{"first_name":"Jürgen","full_name":"Schmidhuber, Jürgen","last_name":"Schmidhuber"},{"first_name":"Olivier","full_name":"Bachem, Olivier","last_name":"Bachem"}],"status":"public","intvolume":"        32","abstract":[{"text":"A disentangled representation encodes information about the salient factors\r\nof variation in the data independently. Although it is often argued that this\r\nrepresentational format is useful in learning to solve many real-world\r\ndown-stream tasks, there is little empirical evidence that supports this claim.\r\nIn this paper, we conduct a large-scale study that investigates whether\r\ndisentangled representations are more suitable for abstract reasoning tasks.\r\nUsing two new tasks similar to Raven's Progressive Matrices, we evaluate the\r\nusefulness of the representations learned by 360 state-of-the-art unsupervised\r\ndisentanglement models. Based on these representations, we train 3600 abstract\r\nreasoning models and observe that disentangled representations do in fact lead\r\nto better down-stream performance. In particular, they enable quicker learning\r\nusing fewer samples.","lang":"eng"}],"volume":32,"date_updated":"2023-09-12T09:02:43Z","oa":1,"article_processing_charge":"No","arxiv":1,"publication":"Advances in Neural Information Processing Systems","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","quality_controlled":"1","_id":"14193","extern":"1","publication_identifier":{"isbn":["9781713807933"]},"date_published":"2019-05-29T00:00:00Z","month":"05","year":"2019","language":[{"iso":"eng"}],"title":"Are disentangled representations helpful for abstract visual reasoning?","external_id":{"arxiv":["1905.12506"]},"department":[{"_id":"FrLo"}],"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.1905.12506","open_access":"1"}],"date_created":"2023-08-22T14:09:53Z","conference":{"start_date":"2019-12-08","location":"Vancouver, Canada","end_date":"2019-12-14","name":"NeurIPS: Neural Information Processing Systems"}},{"date_updated":"2023-09-12T09:37:22Z","volume":32,"oa":1,"article_processing_charge":"No","arxiv":1,"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","quality_controlled":"1","oa_version":"Preprint","_id":"14197","extern":"1","publication_identifier":{"isbn":["9781713807933"]},"publication_status":"published","citation":{"apa":"Locatello, F., Abbati, G., Rainforth, T., Bauer, S., Schölkopf, B., &#38; Bachem, O. (2019). On the fairness of disentangled representations. In <i>Advances in Neural Information Processing Systems</i> (Vol. 32, pp. 14611–14624). Vancouver, Canada.","ieee":"F. Locatello, G. Abbati, T. Rainforth, S. Bauer, B. Schölkopf, and O. Bachem, “On the fairness of disentangled representations,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2019, vol. 32, pp. 14611–14624.","chicago":"Locatello, Francesco, Gabriele Abbati, Tom Rainforth, Stefan Bauer, Bernhard Schölkopf, and Olivier Bachem. “On the Fairness of Disentangled Representations.” In <i>Advances in Neural Information Processing Systems</i>, 32:14611–14624, 2019.","ama":"Locatello F, Abbati G, Rainforth T, Bauer S, Schölkopf B, Bachem O. On the fairness of disentangled representations. In: <i>Advances in Neural Information Processing Systems</i>. Vol 32. ; 2019:14611–14624.","mla":"Locatello, Francesco, et al. “On the Fairness of Disentangled Representations.” <i>Advances in Neural Information Processing Systems</i>, vol. 32, 2019, pp. 14611–14624.","short":"F. Locatello, G. Abbati, T. Rainforth, S. Bauer, B. Schölkopf, O. Bachem, in:, Advances in Neural Information Processing Systems, 2019, pp. 14611–14624.","ista":"Locatello F, Abbati G, Rainforth T, Bauer S, Schölkopf B, Bachem O. 2019. On the fairness of disentangled representations. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 32, 14611–14624."},"author":[{"full_name":"Locatello, Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Gabriele","last_name":"Abbati","full_name":"Abbati, Gabriele"},{"full_name":"Rainforth, Tom","last_name":"Rainforth","first_name":"Tom"},{"first_name":"Stefan","last_name":"Bauer","full_name":"Bauer, Stefan"},{"first_name":"Bernhard","last_name":"Schölkopf","full_name":"Schölkopf, Bernhard"},{"first_name":"Olivier","last_name":"Bachem","full_name":"Bachem, Olivier"}],"abstract":[{"text":"Recently there has been a significant interest in learning disentangled\r\nrepresentations, as they promise increased interpretability, generalization to\r\nunseen scenarios and faster learning on downstream tasks. In this paper, we\r\ninvestigate the usefulness of different notions of disentanglement for\r\nimproving the fairness of downstream prediction tasks based on representations.\r\nWe consider the setting where the goal is to predict a target variable based on\r\nthe learned representation of high-dimensional observations (such as images)\r\nthat depend on both the target variable and an \\emph{unobserved} sensitive\r\nvariable. We show that in this setting both the optimal and empirical\r\npredictions can be unfair, even if the target variable and the sensitive\r\nvariable are independent. Analyzing the representations of more than\r\n\\num{12600} trained state-of-the-art disentangled models, we observe that\r\nseveral disentanglement scores are consistently correlated with increased\r\nfairness, suggesting that disentanglement may be a useful property to encourage\r\nfairness when sensitive variables are not observed.","lang":"eng"}],"main_file_link":[{"url":"https://arxiv.org/abs/1905.13662","open_access":"1"}],"year":"2019","title":"On the fairness of disentangled representations","external_id":{"arxiv":["1905.13662"]},"page":"14611–14624","publication":"Advances in Neural Information Processing Systems","type":"conference","day":"08","status":"public","intvolume":"        32","department":[{"_id":"FrLo"}],"date_created":"2023-08-22T14:12:28Z","conference":{"start_date":"2019-12-08","end_date":"2019-12-14","name":"NeurIPS: Neural Information Processing Systems","location":"Vancouver, Canada"},"date_published":"2019-12-08T00:00:00Z","month":"12","language":[{"iso":"eng"}],"scopus_import":"1"}]