[{"abstract":[{"text":"A natural approach to generative modeling of videos is to represent them as a composition of moving objects. Recent works model a set of 2D sprites over a slowly-varying background, but without considering the underlying 3D scene that\r\ngives rise to them. We instead propose to model a video as the view seen while moving through a scene with multiple 3D objects and a 3D background. Our model is trained from monocular videos without any supervision, yet learns to\r\ngenerate coherent 3D scenes containing several moving objects. We conduct detailed experiments on two datasets, going beyond the visual complexity supported by state-of-the-art generative approaches. We evaluate our method on\r\ndepth-prediction and 3D object detection---tasks which cannot be addressed by those earlier works---and show it out-performs them even on 2D instance segmentation and tracking.","lang":"eng"}],"oa_version":"Preprint","publication_status":"published","page":"3106–3117","day":"07","acknowledged_ssus":[{"_id":"ScienComp"}],"status":"public","date_updated":"2023-04-25T09:49:58Z","external_id":{"arxiv":["2007.06705"]},"title":"Unsupervised object-centric video generation and decomposition in 3D","publication":"34th Conference on Neural Information Processing Systems","main_file_link":[{"url":"https://arxiv.org/abs/2007.06705","open_access":"1"}],"arxiv":1,"citation":{"ieee":"P. M. Henderson and C. Lampert, “Unsupervised object-centric video generation and decomposition in 3D,” in <i>34th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2020, vol. 33, pp. 3106–3117.","chicago":"Henderson, Paul M, and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” In <i>34th Conference on Neural Information Processing Systems</i>, 33:3106–3117. Curran Associates, 2020.","ama":"Henderson PM, Lampert C. Unsupervised object-centric video generation and decomposition in 3D. In: <i>34th Conference on Neural Information Processing Systems</i>. Vol 33. Curran Associates; 2020:3106–3117.","apa":"Henderson, P. M., &#38; Lampert, C. (2020). Unsupervised object-centric video generation and decomposition in 3D. In <i>34th Conference on Neural Information Processing Systems</i> (Vol. 33, pp. 3106–3117). Vancouver, Canada: Curran Associates.","short":"P.M. Henderson, C. Lampert, in:, 34th Conference on Neural Information Processing Systems, Curran Associates, 2020, pp. 3106–3117.","mla":"Henderson, Paul M., and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” <i>34th Conference on Neural Information Processing Systems</i>, vol. 33, Curran Associates, 2020, pp. 3106–3117.","ista":"Henderson PM, Lampert C. 2020. Unsupervised object-centric video generation and decomposition in 3D. 34th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 33, 3106–3117."},"date_published":"2020-07-07T00:00:00Z","oa":1,"_id":"8188","author":[{"full_name":"Henderson, Paul M","first_name":"Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425","last_name":"Henderson","orcid":"0000-0002-5198-7445"},{"last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph"}],"type":"conference","publisher":"Curran Associates","year":"2020","language":[{"iso":"eng"}],"volume":33,"quality_controlled":"1","publication_identifier":{"isbn":["9781713829546"]},"month":"07","intvolume":"        33","article_processing_charge":"No","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","conference":{"end_date":"2020-12-12","start_date":"2020-12-06","name":"NeurIPS: Neural Information Processing Systems","location":"Vancouver, Canada"},"department":[{"_id":"ChLa"}],"acknowledgement":"This research was supported by the Scientific Service Units (SSU) of IST Austria through resources\r\nprovided by Scientific Computing (SciComp). PH is employed part-time by Blackford Analysis, but\r\nthey did not support this project in any way.","date_created":"2020-07-31T16:59:19Z"},{"page":"11525-11538","date_published":"2020-01-01T00:00:00Z","publication_status":"published","oa_version":"Preprint","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"}],"author":[{"first_name":"Francesco","full_name":"Locatello, Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","orcid":"0000-0002-4850-0683"},{"last_name":"Weissenborn","first_name":"Dirk","full_name":"Weissenborn, Dirk"},{"first_name":"Thomas","full_name":"Unterthiner, Thomas","last_name":"Unterthiner"},{"first_name":"Aravindh","full_name":"Mahendran, Aravindh","last_name":"Mahendran"},{"first_name":"Georg","full_name":"Heigold, Georg","last_name":"Heigold"},{"last_name":"Uszkoreit","first_name":"Jakob","full_name":"Uszkoreit, Jakob"},{"full_name":"Dosovitskiy, Alexey","first_name":"Alexey","last_name":"Dosovitskiy"},{"first_name":"Thomas","full_name":"Kipf, Thomas","last_name":"Kipf"}],"oa":1,"_id":"14326","status":"public","type":"conference","language":[{"iso":"eng"}],"date_updated":"2023-09-13T12:19:19Z","publisher":"Curran Associates","year":"2020","publication":"Advances in Neural Information Processing Systems","external_id":{"arxiv":["2006.15055"]},"volume":33,"title":"Object-centric learning with slot attention","quality_controlled":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2006.15055"}],"publication_identifier":{"isbn":["9781713829546"]},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","article_processing_charge":"No","extern":"1","conference":{"location":"Virtual","name":"NeurIPS: Neural Information Processing Systems","start_date":"2020-12-06","end_date":"2020-12-12"},"intvolume":"        33","date_created":"2023-09-13T12:03:46Z","arxiv":1,"citation":{"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.","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.","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.","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.","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.","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."},"department":[{"_id":"FrLo"}]},{"date_updated":"2023-02-23T14:03:03Z","status":"public","oa_version":"Published Version","publication_status":"published","page":"22361-22372","day":"06","ec_funded":1,"abstract":[{"lang":"eng","text":"The ability to leverage large-scale hardware parallelism has been one of the key enablers of the accelerated recent progress in machine learning. Consequently, there has been considerable effort invested into developing efficient parallel variants of classic machine learning algorithms. However, despite the wealth of knowledge on parallelization, some classic machine learning algorithms often prove hard to parallelize efficiently while maintaining convergence. In this paper, we focus on efficient parallel algorithms for the key machine learning task of inference on graphical models, in particular on the fundamental belief propagation algorithm. We address the challenge of efficiently parallelizing this classic paradigm by showing how to leverage scalable relaxed schedulers in this context. We present an extensive empirical study, showing that our approach outperforms previous parallel belief propagation implementations both in terms of scalability and in terms of wall-clock convergence time, on a range of practical applications."}],"project":[{"_id":"268A44D6-B435-11E9-9278-68D0E5697425","name":"Elastic Coordination for Scalable Machine Learning","call_identifier":"H2020","grant_number":"805223"}],"citation":{"chicago":"Aksenov, Vitaly, Dan-Adrian Alistarh, and Janne Korhonen. “Scalable Belief Propagation via Relaxed Scheduling.” In <i>Advances in Neural Information Processing Systems</i>, 33:22361–72. Curran Associates, 2020.","ieee":"V. Aksenov, D.-A. Alistarh, and J. Korhonen, “Scalable belief propagation via relaxed scheduling,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2020, vol. 33, pp. 22361–22372.","ista":"Aksenov V, Alistarh D-A, Korhonen J. 2020. Scalable belief propagation via relaxed scheduling. Advances in Neural Information Processing Systems. NeurIPS: Conference on Neural Information Processing Systems vol. 33, 22361–22372.","apa":"Aksenov, V., Alistarh, D.-A., &#38; Korhonen, J. (2020). Scalable belief propagation via relaxed scheduling. In <i>Advances in Neural Information Processing Systems</i> (Vol. 33, pp. 22361–22372). Vancouver, Canada: Curran Associates.","mla":"Aksenov, Vitaly, et al. “Scalable Belief Propagation via Relaxed Scheduling.” <i>Advances in Neural Information Processing Systems</i>, vol. 33, Curran Associates, 2020, pp. 22361–72.","short":"V. Aksenov, D.-A. Alistarh, J. Korhonen, in:, Advances in Neural Information Processing Systems, Curran Associates, 2020, pp. 22361–22372.","ama":"Aksenov V, Alistarh D-A, Korhonen J. Scalable belief propagation via relaxed scheduling. In: <i>Advances in Neural Information Processing Systems</i>. Vol 33. Curran Associates; 2020:22361-22372."},"arxiv":1,"scopus_import":"1","main_file_link":[{"url":"https://proceedings.neurips.cc/paper/2020/hash/fdb2c3bab9d0701c4a050a4d8d782c7f-Abstract.html","open_access":"1"}],"publication":"Advances in Neural Information Processing Systems","external_id":{"arxiv":["2002.11505"]},"title":"Scalable belief propagation via relaxed scheduling","language":[{"iso":"eng"}],"publisher":"Curran Associates","year":"2020","type":"conference","author":[{"first_name":"Vitaly","full_name":"Aksenov, Vitaly","last_name":"Aksenov"},{"first_name":"Dan-Adrian","full_name":"Alistarh, Dan-Adrian","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","last_name":"Alistarh","orcid":"0000-0003-3650-940X"},{"full_name":"Korhonen, Janne","first_name":"Janne","id":"C5402D42-15BC-11E9-A202-CA2BE6697425","last_name":"Korhonen"}],"oa":1,"_id":"9631","date_published":"2020-12-06T00:00:00Z","date_created":"2021-07-04T22:01:26Z","department":[{"_id":"DaAl"}],"acknowledgement":"We thank Marco Mondelli for discussions related to LDPC decoding, and Giorgi Nadiradze for discussions on analysis of relaxed schedulers. This project has received funding from the European Research Council (ERC) under the European\r\nUnion’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML).","user_id":"6785fbc1-c503-11eb-8a32-93094b40e1cf","article_processing_charge":"No","conference":{"name":"NeurIPS: Conference on Neural Information Processing Systems","start_date":"2020-12-06","end_date":"2020-12-12","location":"Vancouver, Canada"},"intvolume":"        33","month":"12","quality_controlled":"1","publication_identifier":{"isbn":["9781713829546"],"issn":["10495258"]},"volume":33},{"scopus_import":"1","arxiv":1,"citation":{"ama":"Singh SP, Alistarh D-A. WoodFisher: Efficient second-order approximation for neural network compression. In: <i>Advances in Neural Information Processing Systems</i>. Vol 33. Curran Associates; 2020:18098-18109.","short":"S.P. Singh, D.-A. Alistarh, in:, Advances in Neural Information Processing Systems, Curran Associates, 2020, pp. 18098–18109.","ista":"Singh SP, Alistarh D-A. 2020. WoodFisher: Efficient second-order approximation for neural network compression. Advances in Neural Information Processing Systems. NeurIPS: Conference on Neural Information Processing Systems vol. 33, 18098–18109.","mla":"Singh, Sidak Pal, and Dan-Adrian Alistarh. “WoodFisher: Efficient Second-Order Approximation for Neural Network Compression.” <i>Advances in Neural Information Processing Systems</i>, vol. 33, Curran Associates, 2020, pp. 18098–109.","apa":"Singh, S. P., &#38; Alistarh, D.-A. (2020). WoodFisher: Efficient second-order approximation for neural network compression. In <i>Advances in Neural Information Processing Systems</i> (Vol. 33, pp. 18098–18109). Vancouver, Canada: Curran Associates.","ieee":"S. P. Singh and D.-A. Alistarh, “WoodFisher: Efficient second-order approximation for neural network compression,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2020, vol. 33, pp. 18098–18109.","chicago":"Singh, Sidak Pal, and Dan-Adrian Alistarh. “WoodFisher: Efficient Second-Order Approximation for Neural Network Compression.” In <i>Advances in Neural Information Processing Systems</i>, 33:18098–109. Curran Associates, 2020."},"project":[{"grant_number":"805223","call_identifier":"H2020","name":"Elastic Coordination for Scalable Machine Learning","_id":"268A44D6-B435-11E9-9278-68D0E5697425"}],"publication":"Advances in Neural Information Processing Systems","title":"WoodFisher: Efficient second-order approximation for neural network compression","external_id":{"arxiv":["2004.14340"]},"main_file_link":[{"url":"https://proceedings.neurips.cc/paper/2020/hash/d1ff1ec86b62cd5f3903ff19c3a326b2-Abstract.html","open_access":"1"}],"status":"public","date_updated":"2023-02-23T14:03:06Z","day":"06","page":"18098-18109","oa_version":"Published Version","publication_status":"published","abstract":[{"text":"Second-order information, in the form of Hessian- or Inverse-Hessian-vector products, is a fundamental tool for solving optimization problems. Recently, there has been significant interest in utilizing this information in the context of deep\r\nneural networks; however, relatively little is known about the quality of existing approximations in this context. Our work examines this question, identifies issues with existing approaches, and proposes a method called WoodFisher to compute a faithful and efficient estimate of the inverse Hessian. Our main application is to neural network compression, where we build on the classic Optimal Brain Damage/Surgeon framework. We demonstrate that WoodFisher significantly outperforms popular state-of-the-art methods for oneshot pruning. Further, even when iterative, gradual pruning is allowed, our method results in a gain in test accuracy over the state-of-the-art approaches, for standard image classification datasets such as ImageNet ILSVRC. We examine how our method can be extended to take into account first-order information, as well as\r\nillustrate its ability to automatically set layer-wise pruning thresholds and perform compression in the limited-data regime. The code is available at the following link, https://github.com/IST-DASLab/WoodFisher.","lang":"eng"}],"ec_funded":1,"conference":{"end_date":"2020-12-12","start_date":"2020-12-06","name":"NeurIPS: Conference on Neural Information Processing Systems","location":"Vancouver, Canada"},"user_id":"6785fbc1-c503-11eb-8a32-93094b40e1cf","article_processing_charge":"No","intvolume":"        33","date_created":"2021-07-04T22:01:26Z","acknowledgement":"This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML). Also, we would like to thank Alexander Shevchenko, Alexandra Peste, and other members of the group for fruitful discussions.","department":[{"_id":"DaAl"},{"_id":"ToHe"}],"volume":33,"month":"12","publication_identifier":{"isbn":["9781713829546"],"issn":["10495258"]},"quality_controlled":"1","type":"conference","language":[{"iso":"eng"}],"year":"2020","publisher":"Curran Associates","date_published":"2020-12-06T00:00:00Z","author":[{"id":"DD138E24-D89D-11E9-9DC0-DEF6E5697425","last_name":"Singh","full_name":"Singh, Sidak Pal","first_name":"Sidak Pal"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","last_name":"Alistarh","orcid":"0000-0003-3650-940X","first_name":"Dan-Adrian","full_name":"Alistarh, Dan-Adrian"}],"_id":"9632","oa":1}]