[{"abstract":[{"text":"Genetically informed, deep-phenotyped biobanks are an important research resource and it is imperative that the most powerful, versatile, and efficient analysis approaches are used. Here, we apply our recently developed Bayesian grouped mixture of regressions model (GMRM) in the UK and Estonian Biobanks and obtain the highest genomic prediction accuracy reported to date across 21 heritable traits. When compared to other approaches, GMRM accuracy was greater than annotation prediction models run in the LDAK or LDPred-funct software by 15% (SE 7%) and 14% (SE 2%), respectively, and was 18% (SE 3%) greater than a baseline BayesR model without single-nucleotide polymorphism (SNP) markers grouped into minor allele frequency–linkage disequilibrium (MAF-LD) annotation categories. For height, the prediction accuracy R 2 was 47% in a UK Biobank holdout sample, which was 76% of the estimated h SNP 2 . We then extend our GMRM prediction model to provide mixed-linear model association (MLMA) SNP marker estimates for genome-wide association (GWAS) discovery, which increased the independent loci detected to 16,162 in unrelated UK Biobank individuals, compared to 10,550 from BoltLMM and 10,095 from Regenie, a 62 and 65% increase, respectively. The average χ2 value of the leading markers increased by 15.24 (SE 0.41) for every 1% increase in prediction accuracy gained over a baseline BayesR model across the traits. Thus, we show that modeling genetic associations accounting for MAF and LD differences among SNP markers, and incorporating prior knowledge of genomic function, is important for both genomic prediction and discovery in large-scale individual-level studies.","lang":"eng"}],"citation":{"ama":"Orliac E, Trejo Banos D, Ojavee S, et al. Improving genome-wide association discovery and genomic prediction accuracy in biobank data. 2022. doi:10.5061/DRYAD.GTHT76HMZ","apa":"Orliac, E., Trejo Banos, D., Ojavee, S., Läll, K., Mägi, R., Visscher, P., & Robinson, M. R. (2022). Improving genome-wide association discovery and genomic prediction accuracy in biobank data. Dryad. https://doi.org/10.5061/DRYAD.GTHT76HMZ","chicago":"Orliac, Etienne, Daniel Trejo Banos, Sven Ojavee, Kristi Läll, Reedik Mägi, Peter Visscher, and Matthew Richard Robinson. “Improving Genome-Wide Association Discovery and Genomic Prediction Accuracy in Biobank Data.” Dryad, 2022. https://doi.org/10.5061/DRYAD.GTHT76HMZ.","ieee":"E. Orliac et al., “Improving genome-wide association discovery and genomic prediction accuracy in biobank data.” Dryad, 2022.","mla":"Orliac, Etienne, et al. Improving Genome-Wide Association Discovery and Genomic Prediction Accuracy in Biobank Data. Dryad, 2022, doi:10.5061/DRYAD.GTHT76HMZ.","short":"E. Orliac, D. Trejo Banos, S. Ojavee, K. Läll, R. Mägi, P. Visscher, M.R. Robinson, (2022).","ista":"Orliac E, Trejo Banos D, Ojavee S, Läll K, Mägi R, Visscher P, Robinson MR. 2022. Improving genome-wide association discovery and genomic prediction accuracy in biobank data, Dryad, 10.5061/DRYAD.GTHT76HMZ."},"_id":"13064","department":[{"_id":"MaRo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://doi.org/10.5061/dryad.gtht76hmz","open_access":"1"}],"oa_version":"Published Version","publisher":"Dryad","date_published":"2022-09-02T00:00:00Z","related_material":{"record":[{"status":"public","relation":"used_in_publication","id":"11733"}]},"title":"Improving genome-wide association discovery and genomic prediction accuracy in biobank data","ddc":["570"],"tmp":{"image":"/images/cc_0.png","name":"Creative Commons Public Domain Dedication (CC0 1.0)","short":"CC0 (1.0)","legal_code_url":"https://creativecommons.org/publicdomain/zero/1.0/legalcode"},"day":"02","doi":"10.5061/DRYAD.GTHT76HMZ","date_created":"2023-05-23T16:28:13Z","year":"2022","type":"research_data_reference","oa":1,"author":[{"full_name":"Orliac, Etienne","first_name":"Etienne","last_name":"Orliac"},{"last_name":"Trejo Banos","first_name":"Daniel","full_name":"Trejo Banos, Daniel"},{"last_name":"Ojavee","first_name":"Sven","full_name":"Ojavee, Sven"},{"last_name":"Läll","full_name":"Läll, Kristi","first_name":"Kristi"},{"full_name":"Mägi, Reedik","first_name":"Reedik","last_name":"Mägi"},{"full_name":"Visscher, Peter","first_name":"Peter","last_name":"Visscher"},{"first_name":"Matthew Richard","full_name":"Robinson, Matthew Richard","last_name":"Robinson","orcid":"0000-0001-8982-8813","id":"E5D42276-F5DA-11E9-8E24-6303E6697425"}],"month":"09","license":"https://creativecommons.org/publicdomain/zero/1.0/","date_updated":"2023-08-03T12:40:37Z","article_processing_charge":"No","status":"public"},{"status":"public","article_processing_charge":"No","date_updated":"2023-08-04T09:42:10Z","author":[{"full_name":"Koch, Eva","first_name":"Eva","last_name":"Koch"},{"first_name":"Mark","full_name":"Ravinet, Mark","last_name":"Ravinet"},{"first_name":"Anja M","full_name":"Westram, Anja M","last_name":"Westram","id":"3C147470-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0003-1050-4969"},{"last_name":"Jonannesson","full_name":"Jonannesson, Kerstin","first_name":"Kerstin"},{"last_name":"Butlin","full_name":"Butlin, Roger","first_name":"Roger"}],"oa":1,"month":"07","type":"research_data_reference","year":"2022","date_created":"2023-05-23T16:33:12Z","tmp":{"image":"/images/cc_0.png","name":"Creative Commons Public Domain Dedication (CC0 1.0)","short":"CC0 (1.0)","legal_code_url":"https://creativecommons.org/publicdomain/zero/1.0/legalcode"},"day":"28","doi":"10.5061/DRYAD.M905QFV4B","ddc":["570"],"title":"Data from: Genetic architecture of repeated phenotypic divergence in Littorina saxatilis ecotype evolution","related_material":{"record":[{"status":"public","relation":"used_in_publication","id":"12247"}]},"date_published":"2022-07-28T00:00:00Z","publisher":"Dryad","oa_version":"Published Version","main_file_link":[{"url":"https://doi.org/10.5061/dryad.m905qfv4b","open_access":"1"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","department":[{"_id":"NiBa"}],"_id":"13066","citation":{"mla":"Koch, Eva, et al. Data from: Genetic Architecture of Repeated Phenotypic Divergence in Littorina Saxatilis Ecotype Evolution. Dryad, 2022, doi:10.5061/DRYAD.M905QFV4B.","short":"E. Koch, M. Ravinet, A.M. Westram, K. Jonannesson, R. Butlin, (2022).","ieee":"E. Koch, M. Ravinet, A. M. Westram, K. Jonannesson, and R. Butlin, “Data from: Genetic architecture of repeated phenotypic divergence in Littorina saxatilis ecotype evolution.” Dryad, 2022.","ista":"Koch E, Ravinet M, Westram AM, Jonannesson K, Butlin R. 2022. Data from: Genetic architecture of repeated phenotypic divergence in Littorina saxatilis ecotype evolution, Dryad, 10.5061/DRYAD.M905QFV4B.","ama":"Koch E, Ravinet M, Westram AM, Jonannesson K, Butlin R. Data from: Genetic architecture of repeated phenotypic divergence in Littorina saxatilis ecotype evolution. 2022. doi:10.5061/DRYAD.M905QFV4B","apa":"Koch, E., Ravinet, M., Westram, A. M., Jonannesson, K., & Butlin, R. (2022). Data from: Genetic architecture of repeated phenotypic divergence in Littorina saxatilis ecotype evolution. Dryad. https://doi.org/10.5061/DRYAD.M905QFV4B","chicago":"Koch, Eva, Mark Ravinet, Anja M Westram, Kerstin Jonannesson, and Roger Butlin. “Data from: Genetic Architecture of Repeated Phenotypic Divergence in Littorina Saxatilis Ecotype Evolution.” Dryad, 2022. https://doi.org/10.5061/DRYAD.M905QFV4B."},"abstract":[{"text":"Chromosomal inversions have been shown to play a major role in local adaptation by suppressing recombination between alternative arrangements and maintaining beneficial allele combinations. However, so far, their importance relative to the remaining genome remains largely unknown. Understanding the genetic architecture of adaptation requires better estimates of how loci of different effect sizes contribute to phenotypic variation. Here, we used three Swedish islands where the marine snail Littorina saxatilis has repeatedly evolved into two distinct ecotypes along a habitat transition. We estimated the contribution of inversion polymorphisms to phenotypic divergence while controlling for polygenic effects in the remaining genome using a quantitative genetics framework. We confirmed the importance of inversions but showed that contributions of loci outside inversions are of similar magnitude, with variable proportions dependent on the trait and the population. Some inversions showed consistent effects across all sites, whereas others exhibited site-specific effects, indicating that the genomic basis for replicated phenotypic divergence is only partly shared. The contributions of sexual dimorphism as well as environmental factors to phenotypic variation were significant but minor compared to inversions and polygenic background. Overall, this integrated approach provides insight into the multiple mechanisms contributing to parallel phenotypic divergence.","lang":"eng"}]},{"department":[{"_id":"DaAl"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://doi.org/10.5281/zenodo.5813846","open_access":"1"}],"oa_version":"Published Version","publisher":"Zenodo","date_published":"2022-01-03T00:00:00Z","abstract":[{"text":"The source code for replicating experiments presented in the paper.\r\n\r\nThe implementation of the designed priority schedulers can be found in Galois-2.2.1/include/Galois/WorkList/:\r\nStealingMultiQueue.h is the StealingMultiQueue.\r\nMQOptimized/ contains MQ Optimized variants.\r\n\r\nWe provide images that contain all the dependencies and datasets. Images can be pulled from npostnikova/mq-based-schedulers repository, or downloaded from Zenodo. See readme for more detail.","lang":"eng"}],"citation":{"chicago":"Postnikova, Anastasiia, Nikita Koval, Giorgi Nadiradze, and Dan-Adrian Alistarh. “Multi-Queues Can Be State-of-the-Art Priority Schedulers.” Zenodo, 2022. https://doi.org/10.5281/ZENODO.5733408.","apa":"Postnikova, A., Koval, N., Nadiradze, G., & Alistarh, D.-A. (2022). Multi-queues can be state-of-the-art priority schedulers. Zenodo. https://doi.org/10.5281/ZENODO.5733408","ama":"Postnikova A, Koval N, Nadiradze G, Alistarh D-A. Multi-queues can be state-of-the-art priority schedulers. 2022. doi:10.5281/ZENODO.5733408","ista":"Postnikova A, Koval N, Nadiradze G, Alistarh D-A. 2022. Multi-queues can be state-of-the-art priority schedulers, Zenodo, 10.5281/ZENODO.5733408.","ieee":"A. Postnikova, N. Koval, G. Nadiradze, and D.-A. Alistarh, “Multi-queues can be state-of-the-art priority schedulers.” Zenodo, 2022.","short":"A. Postnikova, N. Koval, G. Nadiradze, D.-A. Alistarh, (2022).","mla":"Postnikova, Anastasiia, et al. Multi-Queues Can Be State-of-the-Art Priority Schedulers. Zenodo, 2022, doi:10.5281/ZENODO.5733408."},"_id":"13076","type":"research_data_reference","author":[{"last_name":"Postnikova","full_name":"Postnikova, Anastasiia","first_name":"Anastasiia"},{"last_name":"Koval","full_name":"Koval, Nikita","first_name":"Nikita","id":"2F4DB10C-F248-11E8-B48F-1D18A9856A87"},{"id":"3279A00C-F248-11E8-B48F-1D18A9856A87","last_name":"Nadiradze","first_name":"Giorgi","full_name":"Nadiradze, Giorgi"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0003-3650-940X","full_name":"Alistarh, Dan-Adrian","first_name":"Dan-Adrian","last_name":"Alistarh"}],"oa":1,"month":"01","date_updated":"2023-08-03T06:48:34Z","article_processing_charge":"No","status":"public","title":"Multi-queues can be state-of-the-art priority schedulers","related_material":{"link":[{"relation":"software","url":"https://github.com/npostnikova/mq-based-schedulers/tree/v1.1"}],"record":[{"relation":"used_in_publication","id":"11180","status":"public"}]},"ddc":["510"],"day":"03","doi":"10.5281/ZENODO.5733408","date_created":"2023-05-23T17:05:40Z","year":"2022"},{"file":[{"creator":"dernst","content_type":"application/pdf","file_id":"13243","date_created":"2023-07-18T06:32:38Z","success":1,"file_name":"2022_PMLR_vanderPlas.pdf","file_size":585135,"checksum":"7530a93ef42e10b4db1e5e4b69796e93","access_level":"open_access","relation":"main_file","date_updated":"2023-07-18T06:32:38Z"}],"oa_version":"Published Version","department":[{"_id":"TiVo"}],"date_published":"2022-12-01T00:00:00Z","publication_status":"published","citation":{"ista":"Van Der Plas TL, Vogels TP, Manohar SG. 2022. Predictive learning enables neural networks to learn complex working memory tasks. Proceedings of Machine Learning Research. vol. 199, 518–531.","ieee":"T. L. Van Der Plas, T. P. Vogels, and S. G. Manohar, “Predictive learning enables neural networks to learn complex working memory tasks,” in Proceedings of Machine Learning Research, 2022, vol. 199, pp. 518–531.","short":"T.L. Van Der Plas, T.P. Vogels, S.G. Manohar, in:, Proceedings of Machine Learning Research, ML Research Press, 2022, pp. 518–531.","mla":"Van Der Plas, Thijs L., et al. “Predictive Learning Enables Neural Networks to Learn Complex Working Memory Tasks.” Proceedings of Machine Learning Research, vol. 199, ML Research Press, 2022, pp. 518–31.","chicago":"Van Der Plas, Thijs L., Tim P Vogels, and Sanjay G. Manohar. “Predictive Learning Enables Neural Networks to Learn Complex Working Memory Tasks.” In Proceedings of Machine Learning Research, 199:518–31. ML Research Press, 2022.","ama":"Van Der Plas TL, Vogels TP, Manohar SG. Predictive learning enables neural networks to learn complex working memory tasks. In: Proceedings of Machine Learning Research. Vol 199. ML Research Press; 2022:518-531.","apa":"Van Der Plas, T. L., Vogels, T. P., & Manohar, S. G. (2022). Predictive learning enables neural networks to learn complex working memory tasks. In Proceedings of Machine Learning Research (Vol. 199, pp. 518–531). ML Research Press."},"quality_controlled":"1","oa":1,"status":"public","intvolume":" 199","date_updated":"2023-07-18T06:36:28Z","day":"01","has_accepted_license":"1","title":"Predictive learning enables neural networks to learn complex working memory tasks","publication_identifier":{"eissn":["2640-3498"]},"year":"2022","publisher":"ML Research Press","project":[{"_id":"0aacfa84-070f-11eb-9043-d7eb2c709234","name":"Learning the shape of synaptic plasticity rules for neuronal architectures and function through machine learning.","grant_number":"819603","call_identifier":"H2020"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","ec_funded":1,"abstract":[{"lang":"eng","text":"Brains are thought to engage in predictive learning - learning to predict upcoming stimuli - to construct an internal model of their environment. This is especially notable for spatial navigation, as first described by Tolman’s latent learning tasks. However, predictive learning has also been observed in sensory cortex, in settings unrelated to spatial navigation. Apart from normative frameworks such as active inference or efficient coding, what could be the utility of learning to predict the patterns of occurrence of correlated stimuli? Here we show that prediction, and thereby the construction of an internal model of sequential stimuli, can bootstrap the learning process of a working memory task in a recurrent neural network. We implemented predictive learning alongside working memory match-tasks, and networks emerged to solve the prediction task first by encoding information across time to predict upcoming stimuli, and then eavesdropped on this solution to solve the matching task. Eavesdropping was most beneficial when neural resources were limited. Hence, predictive learning acts as a general neural mechanism to learn to store sensory information that can later be essential for working memory tasks."}],"_id":"13239","acknowledgement":"The authors would like to thank members of the Vogels lab and Manohar lab, as well as Adam Packer, Andrew Saxe, Stefano Sarao Mannelli and Jacob Bakermans for fruitful discussions and comments on earlier versions of the manuscript.\r\nTLvdP was supported by funding from the Biotechnology and Biological Sciences Research Council (BBSRC) [grant number BB/M011224/1]. TPV was supported by an ERC Consolidator Grant (SYNAPSEEK). SGM was funded by a MRC Clinician Scientist Fellowship MR/P00878X and Leverhulme Grant RPG-2018-310.","author":[{"last_name":"Van Der Plas","first_name":"Thijs L.","full_name":"Van Der Plas, Thijs L."},{"first_name":"Tim P","full_name":"Vogels, Tim P","last_name":"Vogels","orcid":"0000-0003-3295-6181","id":"CB6FF8D2-008F-11EA-8E08-2637E6697425"},{"last_name":"Manohar","full_name":"Manohar, Sanjay G.","first_name":"Sanjay G."}],"month":"12","language":[{"iso":"eng"}],"type":"conference","scopus_import":"1","article_processing_charge":"No","publication":"Proceedings of Machine Learning Research","ddc":["000"],"volume":199,"file_date_updated":"2023-07-18T06:32:38Z","date_created":"2023-07-16T22:01:12Z","page":"518-531"},{"date_created":"2023-07-16T22:01:12Z","file_date_updated":"2023-07-17T11:46:34Z","ddc":["579"],"volume":3,"scopus_import":"1","publication":"Frontiers in Fungal Biology","article_processing_charge":"Yes","type":"journal_article","month":"10","language":[{"iso":"eng"}],"author":[{"full_name":"Ingole, Kishor D.","first_name":"Kishor D.","last_name":"Ingole"},{"first_name":"Nithya","full_name":"Nagarajan, Nithya","last_name":"Nagarajan"},{"full_name":"Uhse, Simon","first_name":"Simon","last_name":"Uhse"},{"full_name":"Giannini, Caterina","first_name":"Caterina","last_name":"Giannini","id":"e3fdddd5-f6e0-11ea-865d-ca99ee6367f4"},{"first_name":"Armin","full_name":"Djamei, Armin","last_name":"Djamei"}],"acknowledgement":"The research leading to these results received funding from the European Research Council under the European Union’s Seventh Framework Programme ERC-2013-STG (grant agreement: 335691), the Austrian Science Fund (I 3033-B22), the Austrian Academy of Sciences, and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy EXC-2070-390732324 (PhenoRob) and DFG grant (DJ 64/5-1).\r\nWe would like to thank the GMI/IMBA/IMP core facilities for their excellent technical support. We would like to acknowledge Dr. Sinéad A. O’Sullivan from DZNE, University of Bonn for providing anti-GFP antibodies. The authors are thankful to the Excellence University of Bonn for providing infrastructure and instrumentation facilities at the INRES-Plant Pathology department.","_id":"13240","abstract":[{"lang":"eng","text":"Ustilago maydis is a biotrophic phytopathogenic fungus that causes corn smut disease. As a well-established model system, U. maydis is genetically fully accessible with large omics datasets available and subject to various biological questions ranging from DNA-repair, RNA-transport, and protein secretion to disease biology. For many genetic approaches, tight control of transgene regulation is important. Here we established an optimised version of the Tetracycline-ON (TetON) system for U. maydis. We demonstrate the Tetracycline concentration-dependent expression of fluorescent protein transgenes and the system’s suitability for the induced expression of the toxic protein BCL2 Associated X-1 (Bax1). The Golden Gate compatible vector system contains a native minimal promoter from the mating factor a-1 encoding gene, mfa with ten copies of the tet-regulated operator (tetO) and a codon optimised Tet-repressor (tetR*) which is translationally fused to the native transcriptional corepressor Mql1 (UMAG_05501). The metabolism-independent transcriptional regulator system is functional both, in liquid culture as well as on solid media in the presence of the inducer and can become a useful tool for toxin-antitoxin studies, identification of antifungal proteins, and to study functions of toxic gene products in Ustilago maydis."}],"publisher":"Frontiers Media","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","year":"2022","day":"19","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","image":"/images/cc_by.png"},"doi":"10.3389/ffunb.2022.1029114","article_number":"1029114","has_accepted_license":"1","title":"Tetracycline-controlled (TetON) gene expression system for the smut fungus Ustilago maydis","publication_identifier":{"eissn":["2673-6128"]},"intvolume":" 3","status":"public","date_updated":"2024-03-06T14:01:57Z","oa":1,"quality_controlled":"1","publication_status":"published","citation":{"chicago":"Ingole, Kishor D., Nithya Nagarajan, Simon Uhse, Caterina Giannini, and Armin Djamei. “Tetracycline-Controlled (TetON) Gene Expression System for the Smut Fungus Ustilago Maydis.” Frontiers in Fungal Biology. Frontiers Media, 2022. https://doi.org/10.3389/ffunb.2022.1029114.","ama":"Ingole KD, Nagarajan N, Uhse S, Giannini C, Djamei A. Tetracycline-controlled (TetON) gene expression system for the smut fungus Ustilago maydis. Frontiers in Fungal Biology. 2022;3. doi:10.3389/ffunb.2022.1029114","apa":"Ingole, K. D., Nagarajan, N., Uhse, S., Giannini, C., & Djamei, A. (2022). Tetracycline-controlled (TetON) gene expression system for the smut fungus Ustilago maydis. Frontiers in Fungal Biology. Frontiers Media. https://doi.org/10.3389/ffunb.2022.1029114","ista":"Ingole KD, Nagarajan N, Uhse S, Giannini C, Djamei A. 2022. Tetracycline-controlled (TetON) gene expression system for the smut fungus Ustilago maydis. Frontiers in Fungal Biology. 3, 1029114.","short":"K.D. Ingole, N. Nagarajan, S. Uhse, C. Giannini, A. Djamei, Frontiers in Fungal Biology 3 (2022).","mla":"Ingole, Kishor D., et al. “Tetracycline-Controlled (TetON) Gene Expression System for the Smut Fungus Ustilago Maydis.” Frontiers in Fungal Biology, vol. 3, 1029114, Frontiers Media, 2022, doi:10.3389/ffunb.2022.1029114.","ieee":"K. D. Ingole, N. Nagarajan, S. Uhse, C. Giannini, and A. Djamei, “Tetracycline-controlled (TetON) gene expression system for the smut fungus Ustilago maydis,” Frontiers in Fungal Biology, vol. 3. Frontiers Media, 2022."},"article_type":"original","date_published":"2022-10-19T00:00:00Z","oa_version":"Published Version","file":[{"creator":"dernst","file_id":"13242","content_type":"application/pdf","date_created":"2023-07-17T11:46:34Z","success":1,"file_name":"2023_FrontiersFungalBio_Ingole.pdf","access_level":"open_access","relation":"main_file","checksum":"2254e0119c0749d6f7237084fefcece6","file_size":27966699,"date_updated":"2023-07-17T11:46:34Z"}],"department":[{"_id":"JiFr"}]},{"external_id":{"arxiv":["2102.06004"]},"oa":1,"intvolume":" 171","status":"public","date_updated":"2023-09-26T10:44:37Z","day":"01","publication_identifier":{"eissn":["2640-3498"]},"title":"On the impossibility of fairness-aware learning from corrupted data","year":"2022","oa_version":"Preprint","department":[{"_id":"ChLa"}],"date_published":"2022-12-01T00:00:00Z","publication_status":"published","citation":{"ieee":"N. H. Konstantinov and C. Lampert, “On the impossibility of fairness-aware learning from corrupted data,” in Proceedings of Machine Learning Research, 2022, vol. 171, pp. 59–83.","mla":"Konstantinov, Nikola H., and Christoph Lampert. “On the Impossibility of Fairness-Aware Learning from Corrupted Data.” Proceedings of Machine Learning Research, vol. 171, ML Research Press, 2022, pp. 59–83.","short":"N.H. Konstantinov, C. Lampert, in:, Proceedings of Machine Learning Research, ML Research Press, 2022, pp. 59–83.","ista":"Konstantinov NH, Lampert C. 2022. On the impossibility of fairness-aware learning from corrupted data. Proceedings of Machine Learning Research. vol. 171, 59–83.","ama":"Konstantinov NH, Lampert C. On the impossibility of fairness-aware learning from corrupted data. In: Proceedings of Machine Learning Research. Vol 171. ML Research Press; 2022:59-83.","apa":"Konstantinov, N. H., & Lampert, C. (2022). On the impossibility of fairness-aware learning from corrupted data. In Proceedings of Machine Learning Research (Vol. 171, pp. 59–83). ML Research Press.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “On the Impossibility of Fairness-Aware Learning from Corrupted Data.” In Proceedings of Machine Learning Research, 171:59–83. ML Research Press, 2022."},"quality_controlled":"1","arxiv":1,"type":"conference","language":[{"iso":"eng"}],"month":"12","author":[{"first_name":"Nikola H","full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","full_name":"Lampert, Christoph","last_name":"Lampert"}],"scopus_import":"1","publication":"Proceedings of Machine Learning Research","article_processing_charge":"No","related_material":{"record":[{"relation":"extended_version","id":"10802","status":"public"}]},"volume":171,"date_created":"2023-07-16T22:01:13Z","page":"59-83","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2102.06004"}],"publisher":"ML Research Press","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"text":"Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. Many approaches for training fair models from data have been developed and an implicit assumption about such algorithms is that they are able to recover a fair model, despite potential historical biases in the data. In this work we show a number of impossibility results that indicate that there is no learning algorithm that can recover a fair model when a proportion of the dataset is subject to arbitrary manipulations. Specifically, we prove that there are situations in which an adversary can force any learner to return a biased classifier, with or without degrading accuracy, and that the strength of this bias increases for learning problems with underrepresented protected groups in the data. Our results emphasize on the importance of studying further data corruption models of various strength and of establishing stricter data collection practices for fairness-aware learning.","lang":"eng"}],"acknowledgement":"This paper is a shortened, workshop version of Konstantinov and Lampert (2021),\r\nhttps://arxiv.org/abs/2102.06004. For further results, including an analysis of algorithms achieving the lower bounds from this paper, we refer to the full version.","_id":"13241"},{"publication_status":"published","citation":{"ama":"Dresdner G, Vladarean M-L, Rätsch G, Locatello F, Cevher V, Yurtsever A. Faster one-sample stochastic conditional gradient method for composite convex minimization. In: Proceedings of the 25th International Conference on Artificial Intelligence and Statistics. Vol 151. ML Research Press; 2022:8439-8457.","apa":"Dresdner, G., Vladarean, M.-L., Rätsch, G., Locatello, F., Cevher, V., & Yurtsever, A. (2022). Faster one-sample stochastic conditional gradient method for composite convex minimization. In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (Vol. 151, pp. 8439–8457). Virtual: ML Research Press.","chicago":"Dresdner, Gideon, Maria-Luiza Vladarean, Gunnar Rätsch, Francesco Locatello, Volkan Cevher, and Alp Yurtsever. “ Faster One-Sample Stochastic Conditional Gradient Method for Composite Convex Minimization.” In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics, 151:8439–57. ML Research Press, 2022.","ieee":"G. Dresdner, M.-L. Vladarean, G. Rätsch, F. Locatello, V. Cevher, and A. Yurtsever, “ Faster one-sample stochastic conditional gradient method for composite convex minimization,” in Proceedings of the 25th International Conference on Artificial Intelligence and Statistics, Virtual, 2022, vol. 151, pp. 8439–8457.","short":"G. Dresdner, M.-L. Vladarean, G. Rätsch, F. Locatello, V. Cevher, A. Yurtsever, in:, Proceedings of the 25th International Conference on Artificial Intelligence and Statistics, ML Research Press, 2022, pp. 8439–8457.","mla":"Dresdner, Gideon, et al. “ Faster One-Sample Stochastic Conditional Gradient Method for Composite Convex Minimization.” Proceedings of the 25th International Conference on Artificial Intelligence and Statistics, vol. 151, ML Research Press, 2022, pp. 8439–57.","ista":"Dresdner G, Vladarean M-L, Rätsch G, Locatello F, Cevher V, Yurtsever A. 2022. Faster one-sample stochastic conditional gradient method for composite convex minimization. Proceedings of the 25th International Conference on Artificial Intelligence and Statistics. AISTATS: Conference on Artificial Intelligence and Statistics, PMLR, vol. 151, 8439–8457."},"quality_controlled":"1","arxiv":1,"department":[{"_id":"FrLo"}],"oa_version":"Preprint","date_published":"2022-04-01T00:00:00Z","publication_identifier":{"issn":["2640-3498"]},"title":" Faster one-sample stochastic conditional gradient method for composite convex minimization","day":"01","year":"2022","oa":1,"external_id":{"arxiv":["2202.13212"]},"date_updated":"2023-09-06T10:28:17Z","intvolume":" 151","extern":"1","status":"public","abstract":[{"lang":"eng","text":" We propose a stochastic conditional gradient method (CGM) for minimizing convex finite-sum objectives formed as a sum of smooth and non-smooth terms. Existing CGM variants for this template either suffer from slow convergence rates, or require carefully increasing the batch size over the course of the algorithm’s execution, which leads to computing full gradients. In contrast, the proposed method, equipped with a stochastic average gradient (SAG) estimator, requires only one sample per iteration. Nevertheless, it guarantees fast convergence rates on par with more sophisticated variance reduction techniques. In applications we put special emphasis on problems with a large number of separable constraints. Such problems are prevalent among semidefinite programming (SDP) formulations arising in machine learning and theoretical computer science. We provide numerical experiments on matrix completion, unsupervised clustering, and sparsest-cut SDPs. "}],"alternative_title":["PMLR"],"_id":"14093","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://arxiv.org/abs/2202.13212","open_access":"1"}],"publisher":"ML Research Press","volume":151,"page":"8439-8457","date_created":"2023-08-21T09:27:43Z","type":"conference","language":[{"iso":"eng"}],"author":[{"full_name":"Dresdner, Gideon","first_name":"Gideon","last_name":"Dresdner"},{"first_name":"Maria-Luiza","full_name":"Vladarean, Maria-Luiza","last_name":"Vladarean"},{"last_name":"Rätsch","first_name":"Gunnar","full_name":"Rätsch, Gunnar"},{"first_name":"Francesco","full_name":"Locatello, Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683"},{"full_name":"Cevher, Volkan","first_name":"Volkan","last_name":"Cevher"},{"first_name":"Alp","full_name":"Yurtsever, Alp","last_name":"Yurtsever"}],"conference":{"start_date":"2022-03-28","end_date":"2022-03-30","name":"AISTATS: Conference on Artificial Intelligence and Statistics","location":"Virtual"},"month":"04","publication":"Proceedings of the 25th International Conference on Artificial Intelligence and Statistics","article_processing_charge":"No","scopus_import":"1"},{"publisher":"Neural Information Processing Systems Foundation","main_file_link":[{"url":"https://arxiv.org/abs/2204.04440","open_access":"1"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"text":"We show that deep networks trained to satisfy demographic parity often do so\r\nthrough a form of race or gender awareness, and that the more we force a network\r\nto be fair, the more accurately we can recover race or gender from the internal state\r\nof the network. Based on this observation, we investigate an alternative fairness\r\napproach: we add a second classification head to the network to explicitly predict\r\nthe protected attribute (such as race or gender) alongside the original task. After\r\ntraining the two-headed network, we enforce demographic parity by merging the\r\ntwo heads, creating a network with the same architecture as the original network.\r\nWe establish a close relationship between existing approaches and our approach\r\nby showing (1) that the decisions of a fair classifier are well-approximated by our\r\napproach, and (2) that an unfair and optimally accurate classifier can be recovered\r\nfrom a fair classifier and our second head predicting the protected attribute. We use\r\nour explicit formulation to argue that the existing fairness approaches, just as ours,\r\ndemonstrate disparate treatment and that they are likely to be unlawful in a wide\r\nrange of scenarios under US law.","lang":"eng"}],"_id":"14106","alternative_title":["Advances in Neural Information Processing Systems"],"month":"12","conference":{"end_date":"2022-12-09","start_date":"2022-11-28","name":"NeurIPS: Neural Information Processing Systems","location":"New Orleans, LA, United States"},"language":[{"iso":"eng"}],"author":[{"last_name":"Lohaus","full_name":"Lohaus, Michael","first_name":"Michael"},{"last_name":"Kleindessner","full_name":"Kleindessner, Matthäus","first_name":"Matthäus"},{"first_name":"Krishnaram","full_name":"Kenthapadi, Krishnaram","last_name":"Kenthapadi"},{"orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","full_name":"Locatello, Francesco","first_name":"Francesco","last_name":"Locatello"},{"full_name":"Russell, Chris","first_name":"Chris","last_name":"Russell"}],"type":"conference","scopus_import":"1","article_processing_charge":"No","publication":"36th Conference on Neural Information Processing Systems","volume":35,"date_created":"2023-08-21T12:12:42Z","page":"16548-16562","oa_version":"Preprint","department":[{"_id":"FrLo"}],"date_published":"2022-12-15T00:00:00Z","publication_status":"published","citation":{"chicago":"Lohaus, Michael, Matthäus Kleindessner, Krishnaram Kenthapadi, Francesco Locatello, and Chris Russell. “Are Two Heads the Same as One? Identifying Disparate Treatment in Fair Neural Networks.” In 36th Conference on Neural Information Processing Systems, 35:16548–62. Neural Information Processing Systems Foundation, 2022.","ama":"Lohaus M, Kleindessner M, Kenthapadi K, Locatello F, Russell C. Are two heads the same as one? Identifying disparate treatment in fair neural networks. In: 36th Conference on Neural Information Processing Systems. Vol 35. Neural Information Processing Systems Foundation; 2022:16548-16562.","apa":"Lohaus, M., Kleindessner, M., Kenthapadi, K., Locatello, F., & Russell, C. (2022). Are two heads the same as one? Identifying disparate treatment in fair neural networks. In 36th Conference on Neural Information Processing Systems (Vol. 35, pp. 16548–16562). New Orleans, LA, United States: Neural Information Processing Systems Foundation.","ista":"Lohaus M, Kleindessner M, Kenthapadi K, Locatello F, Russell C. 2022. Are two heads the same as one? Identifying disparate treatment in fair neural networks. 36th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 35, 16548–16562.","ieee":"M. Lohaus, M. Kleindessner, K. Kenthapadi, F. Locatello, and C. Russell, “Are two heads the same as one? Identifying disparate treatment in fair neural networks,” in 36th Conference on Neural Information Processing Systems, New Orleans, LA, United States, 2022, vol. 35, pp. 16548–16562.","mla":"Lohaus, Michael, et al. “Are Two Heads the Same as One? Identifying Disparate Treatment in Fair Neural Networks.” 36th Conference on Neural Information Processing Systems, vol. 35, Neural Information Processing Systems Foundation, 2022, pp. 16548–62.","short":"M. Lohaus, M. Kleindessner, K. Kenthapadi, F. Locatello, C. Russell, in:, 36th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2022, pp. 16548–16562."},"arxiv":1,"quality_controlled":"1","external_id":{"arxiv":["2204.04440"]},"oa":1,"extern":"1","status":"public","intvolume":" 35","date_updated":"2023-09-06T10:29:42Z","day":"15","publication_identifier":{"isbn":["9781713871088"]},"title":"Are two heads the same as one? Identifying disparate treatment in fair neural networks","year":"2022"},{"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2210.12733","open_access":"1"}],"oa_version":"Preprint","department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2022-10-23T00:00:00Z","abstract":[{"lang":"eng","text":"Amodal perception requires inferring the full shape of an object that is partially occluded. This task is particularly challenging on two levels: (1) it requires more information than what is contained in the instant retina or imaging sensor, (2) it is difficult to obtain enough well-annotated amodal labels for supervision. To this end, this paper develops a new framework of\r\nSelf-supervised amodal Video object segmentation (SaVos). Our method efficiently leverages the visual information of video temporal sequences to infer the amodal mask of objects. The key intuition is that the occluded part of an object can be explained away if that part is visible in other frames, possibly deformed as long as the deformation can be reasonably learned.\r\nAccordingly, we derive a novel self-supervised learning paradigm that efficiently utilizes the visible object parts as the supervision to guide the training on videos. In addition to learning type prior to complete masks for known types, SaVos also learns the spatiotemporal prior, which is also useful for the amodal task and could generalize to unseen types. The proposed\r\nframework achieves the state-of-the-art performance on the synthetic amodal segmentation benchmark FISHBOWL and the real world benchmark KINS-Video-Car. Further, it lends itself well to being transferred to novel distributions using test-time adaptation, outperforming existing models even after the transfer to a new distribution."}],"citation":{"ista":"Yao J, Hong Y, Wang C, Xiao T, He T, Locatello F, Wipf D, Fu Y, Zhang Z. 2022. Self-supervised amodal video object segmentation. 36th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems.","ieee":"J. Yao et al., “Self-supervised amodal video object segmentation,” in 36th Conference on Neural Information Processing Systems, New Orleans, LA, United States, 2022.","short":"J. Yao, Y. Hong, C. Wang, T. Xiao, T. He, F. Locatello, D. Wipf, Y. Fu, Z. Zhang, in:, 36th Conference on Neural Information Processing Systems, 2022.","mla":"Yao, Jian, et al. “Self-Supervised Amodal Video Object Segmentation.” 36th Conference on Neural Information Processing Systems, 2022, doi:10.48550/arXiv.2210.12733.","chicago":"Yao, Jian, Yuxin Hong, Chiyu Wang, Tianjun Xiao, Tong He, Francesco Locatello, David Wipf, Yanwei Fu, and Zheng Zhang. “Self-Supervised Amodal Video Object Segmentation.” In 36th Conference on Neural Information Processing Systems, 2022. https://doi.org/10.48550/arXiv.2210.12733.","ama":"Yao J, Hong Y, Wang C, et al. Self-supervised amodal video object segmentation. In: 36th Conference on Neural Information Processing Systems. ; 2022. doi:10.48550/arXiv.2210.12733","apa":"Yao, J., Hong, Y., Wang, C., Xiao, T., He, T., Locatello, F., … Zhang, Z. (2022). Self-supervised amodal video object segmentation. In 36th Conference on Neural Information Processing Systems. New Orleans, LA, United States. https://doi.org/10.48550/arXiv.2210.12733"},"publication_status":"published","_id":"14107","arxiv":1,"external_id":{"arxiv":["2210.12733"]},"type":"conference","author":[{"first_name":"Jian","full_name":"Yao, Jian","last_name":"Yao"},{"first_name":"Yuxin","full_name":"Hong, Yuxin","last_name":"Hong"},{"full_name":"Wang, Chiyu","first_name":"Chiyu","last_name":"Wang"},{"last_name":"Xiao","full_name":"Xiao, Tianjun","first_name":"Tianjun"},{"first_name":"Tong","full_name":"He, Tong","last_name":"He"},{"full_name":"Locatello, Francesco","first_name":"Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"David","full_name":"Wipf, David","last_name":"Wipf"},{"full_name":"Fu, Yanwei","first_name":"Yanwei","last_name":"Fu"},{"full_name":"Zhang, Zheng","first_name":"Zheng","last_name":"Zhang"}],"language":[{"iso":"eng"}],"month":"10","oa":1,"conference":{"name":"NeurIPS: Neural Information Processing Systems","location":"New Orleans, LA, United States","end_date":"2022-12-01","start_date":"2022-11-28"},"status":"public","extern":"1","date_updated":"2023-09-11T09:34:17Z","publication":"36th Conference on Neural Information Processing Systems","article_processing_charge":"No","doi":"10.48550/arXiv.2210.12733","day":"23","title":"Self-supervised amodal video object segmentation","date_created":"2023-08-21T12:13:25Z","year":"2022"},{"date_updated":"2023-09-11T09:19:14Z","extern":"1","status":"public","oa":1,"external_id":{"arxiv":["2203.04913"]},"year":"2022","publication_identifier":{"issn":["1063-6919"],"eissn":["2575-7075"],"isbn":["9781665469470"]},"title":"Leveling down in computer vision: Pareto inefficiencies in fair deep classifiers","day":"01","doi":"10.1109/cvpr52688.2022.01016","date_published":"2022-07-01T00:00:00Z","department":[{"_id":"FrLo"}],"oa_version":"Preprint","quality_controlled":"1","arxiv":1,"publication_status":"published","citation":{"mla":"Zietlow, Dominik, et al. “Leveling down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifiers.” 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 10400–11, doi:10.1109/cvpr52688.2022.01016.","short":"D. Zietlow, M. Lohaus, G. Balakrishnan, M. Kleindessner, F. Locatello, B. Scholkopf, C. Russell, in:, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 10400–10411.","ieee":"D. Zietlow et al., “Leveling down in computer vision: Pareto inefficiencies in fair deep classifiers,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, United States, 2022, pp. 10400–10411.","ista":"Zietlow D, Lohaus M, Balakrishnan G, Kleindessner M, Locatello F, Scholkopf B, Russell C. 2022. Leveling down in computer vision: Pareto inefficiencies in fair deep classifiers. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Conference on Computer Vision and Pattern Recognition, 10400–10411.","ama":"Zietlow D, Lohaus M, Balakrishnan G, et al. Leveling down in computer vision: Pareto inefficiencies in fair deep classifiers. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Institute of Electrical and Electronics Engineers; 2022:10400-10411. doi:10.1109/cvpr52688.2022.01016","apa":"Zietlow, D., Lohaus, M., Balakrishnan, G., Kleindessner, M., Locatello, F., Scholkopf, B., & Russell, C. (2022). Leveling down in computer vision: Pareto inefficiencies in fair deep classifiers. In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10400–10411). New Orleans, LA, United States: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/cvpr52688.2022.01016","chicago":"Zietlow, Dominik, Michael Lohaus, Guha Balakrishnan, Matthaus Kleindessner, Francesco Locatello, Bernhard Scholkopf, and Chris Russell. “Leveling down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifiers.” In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 10400–411. Institute of Electrical and Electronics Engineers, 2022. https://doi.org/10.1109/cvpr52688.2022.01016."},"publication":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition","article_processing_charge":"No","scopus_import":"1","type":"conference","conference":{"location":"New Orleans, LA, United States","name":"CVPR: Conference on Computer Vision and Pattern Recognition","end_date":"2022-06-24","start_date":"2022-06-18"},"author":[{"last_name":"Zietlow","full_name":"Zietlow, Dominik","first_name":"Dominik"},{"last_name":"Lohaus","first_name":"Michael","full_name":"Lohaus, Michael"},{"first_name":"Guha","full_name":"Balakrishnan, Guha","last_name":"Balakrishnan"},{"first_name":"Matthaus","full_name":"Kleindessner, Matthaus","last_name":"Kleindessner"},{"full_name":"Locatello, Francesco","first_name":"Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683"},{"first_name":"Bernhard","full_name":"Scholkopf, Bernhard","last_name":"Scholkopf"},{"last_name":"Russell","full_name":"Russell, Chris","first_name":"Chris"}],"language":[{"iso":"eng"}],"month":"07","page":"10400-10411","date_created":"2023-08-21T12:18:00Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://arxiv.org/abs/2203.04913","open_access":"1"}],"publisher":"Institute of Electrical and Electronics Engineers","_id":"14114","abstract":[{"text":"Algorithmic fairness is frequently motivated in terms of a trade-off in which overall performance is decreased so as to improve performance on disadvantaged groups where the algorithm would otherwise be less accurate. Contrary to this, we find that applying existing fairness approaches to computer vision improve fairness by degrading the performance of classifiers across all groups (with increased degradation on the best performing groups). Extending the bias-variance decomposition for classification to fairness, we theoretically explain why the majority of fairness methods designed for low capacity models should not be used in settings involving high-capacity models, a scenario common to computer vision. We corroborate this analysis with extensive experimental support that shows that many of the fairness heuristics used in computer vision also degrade performance on the most disadvantaged groups. Building on these insights, we propose an adaptive augmentation strategy that, uniquely, of all methods tested, improves performance for the disadvantaged groups.","lang":"eng"}]},{"title":"Neural attentive circuits","volume":35,"day":"14","date_created":"2023-08-22T13:57:27Z","year":"2022","type":"conference","author":[{"first_name":"Nasim","full_name":"Rahaman, Nasim","last_name":"Rahaman"},{"last_name":"Weiss","first_name":"Martin","full_name":"Weiss, Martin"},{"orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","first_name":"Francesco","full_name":"Locatello, Francesco"},{"last_name":"Pal","first_name":"Chris","full_name":"Pal, Chris"},{"last_name":"Bengio","full_name":"Bengio, Yoshua","first_name":"Yoshua"},{"first_name":"Bernhard","full_name":"Schölkopf, Bernhard","last_name":"Schölkopf"},{"first_name":"Li Erran","full_name":"Li, Li Erran","last_name":"Li"},{"last_name":"Ballas","full_name":"Ballas, Nicolas","first_name":"Nicolas"}],"language":[{"iso":"eng"}],"month":"10","oa":1,"conference":{"end_date":"2022-12-01","start_date":"2022-11-29","location":"New Orleans, United States","name":"NeurIPS: Neural Information Processing Systems"},"external_id":{"arxiv":["2210.08031"]},"date_updated":"2023-09-11T09:29:09Z","publication":"36th Conference on Neural Information Processing Systems","article_processing_charge":"No","intvolume":" 35","status":"public","extern":"1","abstract":[{"lang":"eng","text":"Recent work has seen the development of general purpose neural architectures\r\nthat can be trained to perform tasks across diverse data modalities. General\r\npurpose models typically make few assumptions about the underlying\r\ndata-structure and are known to perform well in the large-data regime. At the\r\nsame time, there has been growing interest in modular neural architectures that\r\nrepresent the data using sparsely interacting modules. These models can be more\r\nrobust out-of-distribution, computationally efficient, and capable of\r\nsample-efficient adaptation to new data. However, they tend to make\r\ndomain-specific assumptions about the data, and present challenges in how\r\nmodule behavior (i.e., parameterization) and connectivity (i.e., their layout)\r\ncan be jointly learned. In this work, we introduce a general purpose, yet\r\nmodular neural architecture called Neural Attentive Circuits (NACs) that\r\njointly learns the parameterization and a sparse connectivity of neural modules\r\nwithout using domain knowledge. NACs are best understood as the combination of\r\ntwo systems that are jointly trained end-to-end: one that determines the module\r\nconfiguration and the other that executes it on an input. We demonstrate\r\nqualitatively that NACs learn diverse and meaningful module configurations on\r\nthe NLVR2 dataset without additional supervision. Quantitatively, we show that\r\nby incorporating modularity in this way, NACs improve upon a strong non-modular\r\nbaseline in terms of low-shot adaptation on CIFAR and CUBs dataset by about\r\n10%, and OOD robustness on Tiny ImageNet-R by about 2.5%. Further, we find that\r\nNACs can achieve an 8x speedup at inference time while losing less than 3%\r\nperformance. Finally, we find NACs to yield competitive results on diverse data\r\nmodalities spanning point-cloud classification, symbolic processing and\r\ntext-classification from ASCII bytes, thereby confirming its general purpose\r\nnature."}],"citation":{"ista":"Rahaman N, Weiss M, Locatello F, Pal C, Bengio Y, Schölkopf B, Li LE, Ballas N. 2022. Neural attentive circuits. 36th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 35.","ieee":"N. Rahaman et al., “Neural attentive circuits,” in 36th Conference on Neural Information Processing Systems, New Orleans, United States, 2022, vol. 35.","mla":"Rahaman, Nasim, et al. “Neural Attentive Circuits.” 36th Conference on Neural Information Processing Systems, vol. 35, 2022.","short":"N. Rahaman, M. Weiss, F. Locatello, C. Pal, Y. Bengio, B. Schölkopf, L.E. Li, N. Ballas, in:, 36th Conference on Neural Information Processing Systems, 2022.","chicago":"Rahaman, Nasim, Martin Weiss, Francesco Locatello, Chris Pal, Yoshua Bengio, Bernhard Schölkopf, Li Erran Li, and Nicolas Ballas. “Neural Attentive Circuits.” In 36th Conference on Neural Information Processing Systems, Vol. 35, 2022.","apa":"Rahaman, N., Weiss, M., Locatello, F., Pal, C., Bengio, Y., Schölkopf, B., … Ballas, N. (2022). Neural attentive circuits. In 36th Conference on Neural Information Processing Systems (Vol. 35). New Orleans, United States.","ama":"Rahaman N, Weiss M, Locatello F, et al. Neural attentive circuits. In: 36th Conference on Neural Information Processing Systems. Vol 35. ; 2022."},"publication_status":"published","arxiv":1,"alternative_title":[" Advances in Neural Information Processing Systems"],"_id":"14168","department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2210.08031","open_access":"1"}],"oa_version":"Preprint","date_published":"2022-10-14T00:00:00Z"},{"title":"Generalization and robustness implications in object-centric learning","day":"22","year":"2022","oa":1,"external_id":{"arxiv":["2107.00637"]},"date_updated":"2023-09-11T10:08:14Z","intvolume":" 2022","status":"public","extern":"1","publication_status":"submitted","citation":{"chicago":"Dittadi, Andrea, Samuele Papa, Michele De Vita, Bernhard Schölkopf, Ole Winther, and Francesco Locatello. “Generalization and Robustness Implications in Object-Centric Learning.” In Proceedings of the 39th International Conference on Machine Learning, 2022:5221–85. ML Research Press, n.d.","ama":"Dittadi A, Papa S, Vita MD, Schölkopf B, Winther O, Locatello F. Generalization and robustness implications in object-centric learning. In: Proceedings of the 39th International Conference on Machine Learning. Vol 2022. ML Research Press; :5221-5285.","apa":"Dittadi, A., Papa, S., Vita, M. D., Schölkopf, B., Winther, O., & Locatello, F. (n.d.). Generalization and robustness implications in object-centric learning. In Proceedings of the 39th International Conference on Machine Learning (Vol. 2022, pp. 5221–5285). Baltimore, MD, United States: ML Research Press.","ista":"Dittadi A, Papa S, Vita MD, Schölkopf B, Winther O, Locatello F. Generalization and robustness implications in object-centric learning. Proceedings of the 39th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 2022, 5221–5285.","mla":"Dittadi, Andrea, et al. “Generalization and Robustness Implications in Object-Centric Learning.” Proceedings of the 39th International Conference on Machine Learning, vol. 2022, ML Research Press, pp. 5221–85.","short":"A. Dittadi, S. Papa, M.D. Vita, B. Schölkopf, O. Winther, F. Locatello, in:, Proceedings of the 39th International Conference on Machine Learning, ML Research Press, n.d., pp. 5221–5285.","ieee":"A. Dittadi, S. Papa, M. D. Vita, B. Schölkopf, O. Winther, and F. Locatello, “Generalization and robustness implications in object-centric learning,” in Proceedings of the 39th International Conference on Machine Learning, Baltimore, MD, United States, vol. 2022, pp. 5221–5285."},"quality_controlled":"1","arxiv":1,"department":[{"_id":"FrLo"}],"oa_version":"Preprint","date_published":"2022-07-22T00:00:00Z","volume":2022,"page":"5221-5285","date_created":"2023-08-22T13:59:55Z","type":"conference","author":[{"last_name":"Dittadi","first_name":"Andrea","full_name":"Dittadi, Andrea"},{"full_name":"Papa, Samuele","first_name":"Samuele","last_name":"Papa"},{"last_name":"Vita","first_name":"Michele De","full_name":"Vita, Michele De"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"full_name":"Winther, Ole","first_name":"Ole","last_name":"Winther"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","last_name":"Locatello","first_name":"Francesco","full_name":"Locatello, Francesco"}],"language":[{"iso":"eng"}],"conference":{"name":"International Conference on Machine Learning","location":"Baltimore, MD, United States","end_date":"2022-07-23","start_date":"2022-07-17"},"month":"07","publication":"Proceedings of the 39th International Conference on Machine Learning","article_processing_charge":"No","abstract":[{"lang":"eng","text":"The idea behind object-centric representation learning is that natural scenes can better be modeled as compositions of objects and their relations as opposed to distributed representations. This inductive bias can be injected into neural networks to potentially improve systematic generalization and performance of downstream tasks in scenes with multiple objects. In this paper, we train state-of-the-art unsupervised models on five common multi-object datasets and evaluate segmentation metrics and downstream object property prediction. In addition, we study generalization and robustness by investigating the settings where either a single object is out of distribution -- e.g., having an unseen color, texture, or shape -- or global properties of the scene are altered -- e.g., by occlusions, cropping, or increasing the number of objects. From our experimental study, we find object-centric representations to be useful for\r\ndownstream tasks and generally robust to most distribution shifts affecting objects. However, when the distribution shift affects the input in a less structured manner, robustness in terms of segmentation and downstream task performance may vary significantly across models and distribution shifts. "}],"alternative_title":["PMLR"],"_id":"14170","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2107.00637"}],"publisher":"ML Research Press"},{"day":"22","title":"Score matching enables causal discovery of nonlinear additive noise models","year":"2022","external_id":{"arxiv":["2203.04413"]},"oa":1,"extern":"1","status":"public","intvolume":" 162","date_updated":"2023-09-11T10:14:20Z","citation":{"chicago":"Rolland, Paul, Volkan Cevher, Matthäus Kleindessner, Chris Russel, Bernhard Schölkopf, Dominik Janzing, and Francesco Locatello. “Score Matching Enables Causal Discovery of Nonlinear Additive Noise  Models.” In Proceedings of the 39th International Conference on Machine Learning, 162:18741–53. ML Research Press, 2022.","apa":"Rolland, P., Cevher, V., Kleindessner, M., Russel, C., Schölkopf, B., Janzing, D., & Locatello, F. (2022). Score matching enables causal discovery of nonlinear additive noise  models. In Proceedings of the 39th International Conference on Machine Learning (Vol. 162, pp. 18741–18753). Baltimore, MD, United States: ML Research Press.","ama":"Rolland P, Cevher V, Kleindessner M, et al. Score matching enables causal discovery of nonlinear additive noise  models. In: Proceedings of the 39th International Conference on Machine Learning. Vol 162. ML Research Press; 2022:18741-18753.","ista":"Rolland P, Cevher V, Kleindessner M, Russel C, Schölkopf B, Janzing D, Locatello F. 2022. Score matching enables causal discovery of nonlinear additive noise  models. Proceedings of the 39th International Conference on Machine Learning. International Conference on Machine Learning, PMLR, vol. 162, 18741–18753.","ieee":"P. Rolland et al., “Score matching enables causal discovery of nonlinear additive noise  models,” in Proceedings of the 39th International Conference on Machine Learning, Baltimore, MD, United States, 2022, vol. 162, pp. 18741–18753.","short":"P. Rolland, V. Cevher, M. Kleindessner, C. Russel, B. Schölkopf, D. Janzing, F. Locatello, in:, Proceedings of the 39th International Conference on Machine Learning, ML Research Press, 2022, pp. 18741–18753.","mla":"Rolland, Paul, et al. “Score Matching Enables Causal Discovery of Nonlinear Additive Noise  Models.” Proceedings of the 39th International Conference on Machine Learning, vol. 162, ML Research Press, 2022, pp. 18741–53."},"publication_status":"published","arxiv":1,"quality_controlled":"1","oa_version":"Preprint","department":[{"_id":"FrLo"}],"date_published":"2022-07-22T00:00:00Z","volume":162,"date_created":"2023-08-22T14:00:18Z","page":"18741-18753","language":[{"iso":"eng"}],"month":"07","author":[{"first_name":"Paul","full_name":"Rolland, Paul","last_name":"Rolland"},{"full_name":"Cevher, Volkan","first_name":"Volkan","last_name":"Cevher"},{"full_name":"Kleindessner, Matthäus","first_name":"Matthäus","last_name":"Kleindessner"},{"full_name":"Russel, Chris","first_name":"Chris","last_name":"Russel"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"last_name":"Janzing","first_name":"Dominik","full_name":"Janzing, Dominik"},{"orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","full_name":"Locatello, Francesco","first_name":"Francesco"}],"conference":{"start_date":"2022-07-17","end_date":"2022-07-23","name":"International Conference on Machine Learning","location":"Baltimore, MD, United States"},"type":"conference","article_processing_charge":"No","publication":"Proceedings of the 39th International Conference on Machine Learning","abstract":[{"text":"This paper demonstrates how to recover causal graphs from the score of the\r\ndata distribution in non-linear additive (Gaussian) noise models. Using score\r\nmatching algorithms as a building block, we show how to design a new generation\r\nof scalable causal discovery methods. To showcase our approach, we also propose\r\na new efficient method for approximating the score's Jacobian, enabling to\r\nrecover the causal graph. Empirically, we find that the new algorithm, called\r\nSCORE, is competitive with state-of-the-art causal discovery methods while\r\nbeing significantly faster.","lang":"eng"}],"_id":"14171","alternative_title":["PMLR"],"publisher":"ML Research Press","main_file_link":[{"url":"https://arxiv.org/abs/2203.04413","open_access":"1"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87"},{"abstract":[{"text":"An important component for generalization in machine learning is to uncover underlying latent factors of variation as well as the mechanism through which each factor acts in the world. In this paper, we test whether 17 unsupervised, weakly supervised, and fully supervised representation learning approaches correctly infer the generative factors of variation in simple datasets (dSprites, Shapes3D, MPI3D) from controlled environments, and on our contributed CelebGlow dataset. In contrast to prior robustness work that introduces novel factors of variation during test time, such as blur or other (un)structured noise, we here recompose, interpolate, or extrapolate only existing factors of variation from the training data set (e.g., small and medium-sized objects during training and large objects during testing). Models\r\nthat learn the correct mechanism should be able to generalize to this benchmark. In total, we train and test 2000+ models and observe that all of them struggle to learn the underlying mechanism regardless of supervision signal and architectural bias. Moreover, the generalization capabilities of all tested models drop significantly as we move from artificial datasets towards\r\nmore realistic real-world datasets. Despite their inability to identify the correct mechanism, the models are quite modular as their ability to infer other in-distribution factors remains fairly stable, providing only a single factoris out-of-distribution. These results point to an important yet understudied problem of learning mechanistic models of observations that can facilitate\r\ngeneralization.","lang":"eng"}],"publication_status":"published","citation":{"ista":"Schott L, Kügelgen J von, Träuble F, Gehler P, Russell C, Bethge M, Schölkopf B, Locatello F, Brendel W. 2022. Visual representation learning does not generalize strongly within the  same domain. 10th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","ieee":"L. Schott et al., “Visual representation learning does not generalize strongly within the  same domain,” in 10th International Conference on Learning Representations, Virtual, 2022.","short":"L. Schott, J. von Kügelgen, F. Träuble, P. Gehler, C. Russell, M. Bethge, B. Schölkopf, F. Locatello, W. Brendel, in:, 10th International Conference on Learning Representations, 2022.","mla":"Schott, Lukas, et al. “Visual Representation Learning Does Not Generalize Strongly within the  Same Domain.” 10th International Conference on Learning Representations, 2022.","chicago":"Schott, Lukas, Julius von Kügelgen, Frederik Träuble, Peter Gehler, Chris Russell, Matthias Bethge, Bernhard Schölkopf, Francesco Locatello, and Wieland Brendel. “Visual Representation Learning Does Not Generalize Strongly within the  Same Domain.” In 10th International Conference on Learning Representations, 2022.","ama":"Schott L, Kügelgen J von, Träuble F, et al. Visual representation learning does not generalize strongly within the  same domain. In: 10th International Conference on Learning Representations. ; 2022.","apa":"Schott, L., Kügelgen, J. von, Träuble, F., Gehler, P., Russell, C., Bethge, M., … Brendel, W. (2022). Visual representation learning does not generalize strongly within the  same domain. In 10th International Conference on Learning Representations. Virtual."},"quality_controlled":"1","arxiv":1,"_id":"14172","department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2107.08221","open_access":"1"}],"oa_version":"Preprint","date_published":"2022-04-25T00:00:00Z","title":"Visual representation learning does not generalize strongly within the same domain","day":"25","date_created":"2023-08-22T14:00:50Z","year":"2022","type":"conference","oa":1,"month":"04","language":[{"iso":"eng"}],"author":[{"first_name":"Lukas","full_name":"Schott, Lukas","last_name":"Schott"},{"full_name":"Kügelgen, Julius von","first_name":"Julius von","last_name":"Kügelgen"},{"last_name":"Träuble","full_name":"Träuble, Frederik","first_name":"Frederik"},{"last_name":"Gehler","full_name":"Gehler, Peter","first_name":"Peter"},{"last_name":"Russell","first_name":"Chris","full_name":"Russell, Chris"},{"first_name":"Matthias","full_name":"Bethge, Matthias","last_name":"Bethge"},{"last_name":"Schölkopf","first_name":"Bernhard","full_name":"Schölkopf, Bernhard"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","last_name":"Locatello","first_name":"Francesco","full_name":"Locatello, Francesco"},{"last_name":"Brendel","first_name":"Wieland","full_name":"Brendel, Wieland"}],"conference":{"start_date":"2022-04-25","end_date":"2022-04-29","name":"ICLR: International Conference on Learning Representations","location":"Virtual"},"external_id":{"arxiv":["2107.08221"]},"publication":"10th International Conference on Learning Representations","date_updated":"2023-09-11T09:40:52Z","article_processing_charge":"No","extern":"1","status":"public"},{"_id":"14173","alternative_title":["Advances in Neural Information Processing Systems"],"abstract":[{"lang":"eng","text":"Since out-of-distribution generalization is a generally ill-posed problem, various proxy targets (e.g., calibration, adversarial robustness, algorithmic corruptions, invariance across shifts) were studied across different research programs resulting in different recommendations. While sharing the same aspirational goal, these approaches have never been tested under the same\r\nexperimental conditions on real data. In this paper, we take a unified view of previous work, highlighting message discrepancies that we address empirically, and providing recommendations on how to measure the robustness of a model and how to improve it. To this end, we collect 172 publicly available dataset pairs for training and out-of-distribution evaluation of accuracy, calibration error, adversarial attacks, environment invariance, and synthetic corruptions. We fine-tune over 31k networks, from nine different architectures in the many- and\r\nfew-shot setting. Our findings confirm that in- and out-of-distribution accuracies tend to increase jointly, but show that their relation is largely dataset-dependent, and in general more nuanced and more complex than posited by previous, smaller scale studies."}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publisher":"Neural Information Processing Systems Foundation","main_file_link":[{"url":"https://arxiv.org/abs/2207.09239","open_access":"1"}],"page":"7181-7198","date_created":"2023-08-22T14:01:13Z","volume":35,"article_processing_charge":"No","publication":"36th Conference on Neural Information Processing Systems","scopus_import":"1","month":"12","language":[{"iso":"eng"}],"author":[{"full_name":"Wenzel, Florian","first_name":"Florian","last_name":"Wenzel"},{"last_name":"Dittadi","full_name":"Dittadi, Andrea","first_name":"Andrea"},{"last_name":"Gehler","full_name":"Gehler, Peter Vincent","first_name":"Peter Vincent"},{"last_name":"Carl-Johann Simon-Gabriel","first_name":"Carl-Johann Simon-Gabriel","full_name":"Carl-Johann Simon-Gabriel, Carl-Johann Simon-Gabriel"},{"full_name":"Horn, Max","first_name":"Max","last_name":"Horn"},{"full_name":"Zietlow, Dominik","first_name":"Dominik","last_name":"Zietlow"},{"first_name":"David","full_name":"Kernert, David","last_name":"Kernert"},{"full_name":"Russell, Chris","first_name":"Chris","last_name":"Russell"},{"full_name":"Brox, Thomas","first_name":"Thomas","last_name":"Brox"},{"full_name":"Schiele, Bernt","first_name":"Bernt","last_name":"Schiele"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","first_name":"Francesco","full_name":"Locatello, Francesco"}],"conference":{"location":"New Orleans, LA, United States","name":"NeurIPS: Neural Information Processing Systems","end_date":"2022-12-09","start_date":"2022-11-28"},"type":"conference","arxiv":1,"quality_controlled":"1","publication_status":"published","citation":{"ista":"Wenzel F, Dittadi A, Gehler PV, Carl-Johann Simon-Gabriel C-JS-G, Horn M, Zietlow D, Kernert D, Russell C, Brox T, Schiele B, Schölkopf B, Locatello F. 2022. Assaying out-of-distribution generalization in transfer learning. 36th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 35, 7181–7198.","mla":"Wenzel, Florian, et al. “Assaying Out-of-Distribution Generalization in Transfer Learning.” 36th Conference on Neural Information Processing Systems, vol. 35, Neural Information Processing Systems Foundation, 2022, pp. 7181–98.","short":"F. Wenzel, A. Dittadi, P.V. Gehler, C.-J.S.-G. Carl-Johann Simon-Gabriel, M. Horn, D. Zietlow, D. Kernert, C. Russell, T. Brox, B. Schiele, B. Schölkopf, F. Locatello, in:, 36th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2022, pp. 7181–7198.","ieee":"F. Wenzel et al., “Assaying out-of-distribution generalization in transfer learning,” in 36th Conference on Neural Information Processing Systems, New Orleans, LA, United States, 2022, vol. 35, pp. 7181–7198.","chicago":"Wenzel, Florian, Andrea Dittadi, Peter Vincent Gehler, Carl-Johann Simon-Gabriel Carl-Johann Simon-Gabriel, Max Horn, Dominik Zietlow, David Kernert, et al. “Assaying Out-of-Distribution Generalization in Transfer Learning.” In 36th Conference on Neural Information Processing Systems, 35:7181–98. Neural Information Processing Systems Foundation, 2022.","ama":"Wenzel F, Dittadi A, Gehler PV, et al. Assaying out-of-distribution generalization in transfer learning. In: 36th Conference on Neural Information Processing Systems. Vol 35. Neural Information Processing Systems Foundation; 2022:7181-7198.","apa":"Wenzel, F., Dittadi, A., Gehler, P. V., Carl-Johann Simon-Gabriel, C.-J. S.-G., Horn, M., Zietlow, D., … Locatello, F. (2022). Assaying out-of-distribution generalization in transfer learning. In 36th Conference on Neural Information Processing Systems (Vol. 35, pp. 7181–7198). New Orleans, LA, United States: Neural Information Processing Systems Foundation."},"date_published":"2022-12-15T00:00:00Z","department":[{"_id":"FrLo"}],"oa_version":"Preprint","year":"2022","publication_identifier":{"isbn":["9781713871088"]},"title":"Assaying out-of-distribution generalization in transfer learning","day":"15","date_updated":"2023-09-06T10:34:43Z","extern":"1","status":"public","intvolume":" 35","oa":1,"external_id":{"arxiv":["2207.09239"]}},{"abstract":[{"lang":"eng","text":"Building sample-efficient agents that generalize out-of-distribution (OOD) in real-world settings remains a fundamental unsolved problem on the path towards achieving higher-level cognition. One particularly promising approach is to begin with low-dimensional, pretrained representations of our world, which should facilitate efficient downstream learning and generalization. By training 240 representations and over 10,000 reinforcement learning (RL) policies on a simulated robotic setup, we evaluate to what extent different properties of\r\npretrained VAE-based representations affect the OOD generalization of downstream agents. We observe that many agents are surprisingly robust to realistic distribution shifts, including the challenging sim-to-real case. In addition, we find that the generalization performance of a simple downstream proxy task reliably predicts the generalization performance of our RL agents\r\nunder a wide range of OOD settings. Such proxy tasks can thus be used to select pretrained representations that will lead to agents that generalize."}],"publication_status":"published","citation":{"ista":"Dittadi A, Träuble F, Wüthrich M, Widmaier F, Gehler P, Winther O, Locatello F, Bachem O, Schölkopf B, Bauer S. 2022. The role of pretrained representations for the OOD generalization of  reinforcement learning agents. 10th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","short":"A. Dittadi, F. Träuble, M. Wüthrich, F. Widmaier, P. Gehler, O. Winther, F. Locatello, O. Bachem, B. Schölkopf, S. Bauer, in:, 10th International Conference on Learning Representations, 2022.","mla":"Dittadi, Andrea, et al. “The Role of Pretrained Representations for the OOD Generalization of  Reinforcement Learning Agents.” 10th International Conference on Learning Representations, 2022.","ieee":"A. Dittadi et al., “The role of pretrained representations for the OOD generalization of  reinforcement learning agents,” in 10th International Conference on Learning Representations, Virtual, 2022.","chicago":"Dittadi, Andrea, Frederik Träuble, Manuel Wüthrich, Felix Widmaier, Peter Gehler, Ole Winther, Francesco Locatello, Olivier Bachem, Bernhard Schölkopf, and Stefan Bauer. “The Role of Pretrained Representations for the OOD Generalization of  Reinforcement Learning Agents.” In 10th International Conference on Learning Representations, 2022.","ama":"Dittadi A, Träuble F, Wüthrich M, et al. The role of pretrained representations for the OOD generalization of  reinforcement learning agents. In: 10th International Conference on Learning Representations. ; 2022.","apa":"Dittadi, A., Träuble, F., Wüthrich, M., Widmaier, F., Gehler, P., Winther, O., … Bauer, S. (2022). The role of pretrained representations for the OOD generalization of  reinforcement learning agents. In 10th International Conference on Learning Representations. Virtual."},"_id":"14174","quality_controlled":"1","arxiv":1,"main_file_link":[{"open_access":"1","url":" https://doi.org/10.48550/arXiv.2107.05686"}],"oa_version":"Preprint","department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2022-04-25T00:00:00Z","day":"25","title":"The role of pretrained representations for the OOD generalization of reinforcement learning agents","date_created":"2023-08-22T14:02:13Z","year":"2022","external_id":{"arxiv":["2107.05686"]},"type":"conference","conference":{"start_date":"2022-04-25","end_date":"2022-04-29","name":"ICLR: International Conference on Learning Representations","location":"Virtual"},"language":[{"iso":"eng"}],"oa":1,"month":"04","author":[{"last_name":"Dittadi","full_name":"Dittadi, Andrea","first_name":"Andrea"},{"first_name":"Frederik","full_name":"Träuble, Frederik","last_name":"Träuble"},{"full_name":"Wüthrich, Manuel","first_name":"Manuel","last_name":"Wüthrich"},{"first_name":"Felix","full_name":"Widmaier, Felix","last_name":"Widmaier"},{"last_name":"Gehler","full_name":"Gehler, Peter","first_name":"Peter"},{"first_name":"Ole","full_name":"Winther, Ole","last_name":"Winther"},{"full_name":"Locatello, Francesco","first_name":"Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683"},{"last_name":"Bachem","full_name":"Bachem, Olivier","first_name":"Olivier"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"},{"last_name":"Bauer","first_name":"Stefan","full_name":"Bauer, Stefan"}],"extern":"1","status":"public","publication":"10th International Conference on Learning Representations","date_updated":"2023-09-11T09:48:36Z","article_processing_charge":"No"},{"year":"2022","date_created":"2023-08-22T14:02:34Z","title":"You mostly walk alone: Analyzing feature attribution in trajectory prediction","day":"25","article_processing_charge":"No","publication":"10th International Conference on Learning Representations","date_updated":"2023-09-11T09:52:20Z","extern":"1","status":"public","author":[{"full_name":"Makansi, Osama","first_name":"Osama","last_name":"Makansi"},{"last_name":"Kügelgen","full_name":"Kügelgen, Julius von","first_name":"Julius von"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco","first_name":"Francesco","last_name":"Locatello"},{"first_name":"Peter","full_name":"Gehler, Peter","last_name":"Gehler"},{"full_name":"Janzing, Dominik","first_name":"Dominik","last_name":"Janzing"},{"full_name":"Brox, Thomas","first_name":"Thomas","last_name":"Brox"},{"last_name":"Schölkopf","full_name":"Schölkopf, Bernhard","first_name":"Bernhard"}],"conference":{"name":"ICLR: International Conference on Learning Representations","location":"Virtual","start_date":"2022-04-25","end_date":"2022-04-29"},"language":[{"iso":"eng"}],"month":"04","oa":1,"type":"conference","external_id":{"arxiv":["2110.05304"]},"arxiv":1,"quality_controlled":"1","_id":"14175","citation":{"chicago":"Makansi, Osama, Julius von Kügelgen, Francesco Locatello, Peter Gehler, Dominik Janzing, Thomas Brox, and Bernhard Schölkopf. “You Mostly Walk Alone: Analyzing Feature Attribution in Trajectory Prediction.” In 10th International Conference on Learning Representations, 2022.","apa":"Makansi, O., Kügelgen, J. von, Locatello, F., Gehler, P., Janzing, D., Brox, T., & Schölkopf, B. (2022). You mostly walk alone: Analyzing feature attribution in trajectory prediction. In 10th International Conference on Learning Representations. Virtual.","ama":"Makansi O, Kügelgen J von, Locatello F, et al. You mostly walk alone: Analyzing feature attribution in trajectory prediction. In: 10th International Conference on Learning Representations. ; 2022.","ista":"Makansi O, Kügelgen J von, Locatello F, Gehler P, Janzing D, Brox T, Schölkopf B. 2022. You mostly walk alone: Analyzing feature attribution in trajectory prediction. 10th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","mla":"Makansi, Osama, et al. “You Mostly Walk Alone: Analyzing Feature Attribution in Trajectory Prediction.” 10th International Conference on Learning Representations, 2022.","short":"O. Makansi, J. von Kügelgen, F. Locatello, P. Gehler, D. Janzing, T. Brox, B. Schölkopf, in:, 10th International Conference on Learning Representations, 2022.","ieee":"O. Makansi et al., “You mostly walk alone: Analyzing feature attribution in trajectory prediction,” in 10th International Conference on Learning Representations, Virtual, 2022."},"publication_status":"published","abstract":[{"lang":"eng","text":"Predicting the future trajectory of a moving agent can be easy when the past trajectory continues smoothly but is challenging when complex interactions with other agents are involved. Recent deep learning approaches for trajectory prediction show promising performance and partially attribute this to successful reasoning about agent-agent interactions. However, it remains unclear which features such black-box models actually learn to use for making predictions. This paper proposes a procedure that quantifies the contributions\r\nof different cues to model performance based on a variant of Shapley values. Applying this procedure to state-of-the-art trajectory prediction methods on standard benchmark datasets shows that they are, in fact, unable to reason about interactions. Instead, the past trajectory of the target is the only feature used for predicting its future. For a task with richer social\r\ninteraction patterns, on the other hand, the tested models do pick up such interactions to a certain extent, as quantified by our feature attribution method. We discuss the limits of the proposed method and its links to causality."}],"date_published":"2022-04-25T00:00:00Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","department":[{"_id":"FrLo"}],"oa_version":"Preprint","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2110.05304"}]},{"type":"conference","conference":{"name":"NeurIPS: Neural Information Processing Systems","location":"New Orleans, LA, United States","end_date":"2022-12-09","start_date":"2022-11-28"},"month":"11","oa":1,"language":[{"iso":"eng"}],"author":[{"last_name":"Rahaman","full_name":"Rahaman, Nasim","first_name":"Nasim"},{"last_name":"Weiss","first_name":"Martin","full_name":"Weiss, Martin"},{"first_name":"Frederik","full_name":"Träuble, Frederik","last_name":"Träuble"},{"orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","full_name":"Locatello, Francesco","last_name":"Locatello"},{"full_name":"Lacoste, Alexandre","first_name":"Alexandre","last_name":"Lacoste"},{"first_name":"Yoshua","full_name":"Bengio, Yoshua","last_name":"Bengio"},{"last_name":"Pal","first_name":"Chris","full_name":"Pal, Chris"},{"full_name":"Li, Li Erran","first_name":"Li Erran","last_name":"Li"},{"first_name":"Bernhard","full_name":"Schölkopf, Bernhard","last_name":"Schölkopf"}],"external_id":{"arxiv":["2211.02348"]},"publication":"36th Conference on Neural Information Processing Systems","date_updated":"2023-09-13T09:35:59Z","article_processing_charge":"No","status":"public","extern":"1","title":"A general purpose neural architecture for geospatial systems","day":"04","date_created":"2023-08-22T14:21:47Z","year":"2022","department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2211.02348"}],"oa_version":"Preprint","date_published":"2022-11-04T00:00:00Z","abstract":[{"lang":"eng","text":"Geospatial Information Systems are used by researchers and Humanitarian Assistance and Disaster Response (HADR) practitioners to support a wide variety of important applications. However, collaboration between these actors is difficult due to the heterogeneous nature of geospatial data modalities (e.g., multi-spectral images of various resolutions, timeseries, weather data) and diversity of tasks (e.g., regression of human activity indicators or detecting forest fires). In this work, we present a roadmap towards the construction of a general-purpose neural architecture (GPNA) with a geospatial inductive bias, pre-trained on large amounts of unlabelled earth observation data in a self-supervised manner. We envision how such a model may facilitate cooperation between members of the community. We show preliminary results on the first step of the roadmap, where we instantiate an architecture that can process a wide variety of geospatial data modalities and demonstrate that it can achieve competitive performance with domain-specific architectures on tasks relating to the U.N.'s Sustainable Development Goals."}],"publication_status":"submitted","citation":{"chicago":"Rahaman, Nasim, Martin Weiss, Frederik Träuble, Francesco Locatello, Alexandre Lacoste, Yoshua Bengio, Chris Pal, Li Erran Li, and Bernhard Schölkopf. “A General Purpose Neural Architecture for Geospatial Systems.” In 36th Conference on Neural Information Processing Systems, n.d.","apa":"Rahaman, N., Weiss, M., Träuble, F., Locatello, F., Lacoste, A., Bengio, Y., … Schölkopf, B. (n.d.). A general purpose neural architecture for geospatial systems. In 36th Conference on Neural Information Processing Systems. New Orleans, LA, United States.","ama":"Rahaman N, Weiss M, Träuble F, et al. A general purpose neural architecture for geospatial systems. In: 36th Conference on Neural Information Processing Systems.","ista":"Rahaman N, Weiss M, Träuble F, Locatello F, Lacoste A, Bengio Y, Pal C, Li LE, Schölkopf B. A general purpose neural architecture for geospatial systems. 36th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems.","short":"N. Rahaman, M. Weiss, F. Träuble, F. Locatello, A. Lacoste, Y. Bengio, C. Pal, L.E. Li, B. Schölkopf, in:, 36th Conference on Neural Information Processing Systems, n.d.","mla":"Rahaman, Nasim, et al. “A General Purpose Neural Architecture for Geospatial Systems.” 36th Conference on Neural Information Processing Systems.","ieee":"N. Rahaman et al., “A general purpose neural architecture for geospatial systems,” in 36th Conference on Neural Information Processing Systems, New Orleans, LA, United States."},"quality_controlled":"1","arxiv":1,"_id":"14215"},{"department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2210.01738"}],"oa_version":"Preprint","date_published":"2022-10-04T00:00:00Z","abstract":[{"text":"CLIP proved that aligning visual and language spaces is key to solving many vision tasks without explicit training, but required to train image and text encoders from scratch on a huge dataset. LiT improved this by only training the text encoder and using a pre-trained vision network. In this paper, we show that a common space can be created without any training at all, using single-domain encoders (trained with or without supervision) and a much smaller amount of image-text pairs. Furthermore, our model has unique properties. Most notably, deploying a new version with updated training samples can be done in a matter of seconds. Additionally, the representations in the common space are easily interpretable as every dimension corresponds to the similarity of the input to a unique entry in the multimodal dataset. Experiments on standard zero-shot visual benchmarks demonstrate the typical transfer ability of image-text models. Overall, our method represents a simple yet surprisingly strong baseline for foundation multi-modal models, raising important questions on their data efficiency and on the role of retrieval in machine learning.","lang":"eng"}],"publication_status":"submitted","citation":{"ista":"Norelli A, Fumero M, Maiorca V, Moschella L, Rodolà E, Locatello F. ASIF: Coupled data turns unimodal models to multimodal without training. arXiv, 2210.01738.","mla":"Norelli, Antonio, et al. “ASIF: Coupled Data Turns Unimodal Models to Multimodal without Training.” ArXiv, 2210.01738, doi:10.48550/arXiv.2210.01738.","short":"A. Norelli, M. Fumero, V. Maiorca, L. Moschella, E. Rodolà, F. Locatello, ArXiv (n.d.).","ieee":"A. Norelli, M. Fumero, V. Maiorca, L. Moschella, E. Rodolà, and F. Locatello, “ASIF: Coupled data turns unimodal models to multimodal without training,” arXiv. .","chicago":"Norelli, Antonio, Marco Fumero, Valentino Maiorca, Luca Moschella, Emanuele Rodolà, and Francesco Locatello. “ASIF: Coupled Data Turns Unimodal Models to Multimodal without Training.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2210.01738.","apa":"Norelli, A., Fumero, M., Maiorca, V., Moschella, L., Rodolà, E., & Locatello, F. (n.d.). ASIF: Coupled data turns unimodal models to multimodal without training. arXiv. https://doi.org/10.48550/arXiv.2210.01738","ama":"Norelli A, Fumero M, Maiorca V, Moschella L, Rodolà E, Locatello F. ASIF: Coupled data turns unimodal models to multimodal without training. arXiv. doi:10.48550/arXiv.2210.01738"},"arxiv":1,"_id":"14216","type":"preprint","oa":1,"language":[{"iso":"eng"}],"author":[{"last_name":"Norelli","first_name":"Antonio","full_name":"Norelli, Antonio"},{"first_name":"Marco","full_name":"Fumero, Marco","last_name":"Fumero"},{"full_name":"Maiorca, Valentino","first_name":"Valentino","last_name":"Maiorca"},{"last_name":"Moschella","full_name":"Moschella, Luca","first_name":"Luca"},{"last_name":"Rodolà","full_name":"Rodolà, Emanuele","first_name":"Emanuele"},{"last_name":"Locatello","first_name":"Francesco","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"}],"month":"10","external_id":{"arxiv":["2210.01738"]},"date_updated":"2024-02-12T09:57:14Z","publication":"arXiv","article_processing_charge":"No","status":"public","title":"ASIF: Coupled data turns unimodal models to multimodal without training","article_number":"2210.01738","doi":"10.48550/arXiv.2210.01738","day":"04","date_created":"2023-08-22T14:22:04Z","year":"2022"},{"type":"preprint","month":"01","author":[{"last_name":"Mambelli","full_name":"Mambelli, Davide","first_name":"Davide"},{"last_name":"Träuble","full_name":"Träuble, Frederik","first_name":"Frederik"},{"last_name":"Bauer","first_name":"Stefan","full_name":"Bauer, Stefan"},{"last_name":"Schölkopf","first_name":"Bernhard","full_name":"Schölkopf, Bernhard"},{"orcid":"0000-0002-4850-0683","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","full_name":"Locatello, Francesco","first_name":"Francesco"}],"oa":1,"language":[{"iso":"eng"}],"external_id":{"arxiv":["2201.13388"]},"publication":"arXiv","date_updated":"2023-09-11T11:49:40Z","article_processing_charge":"No","status":"public","extern":"1","title":"Compositional multi-object reinforcement learning with linear relation networks","day":"31","doi":"10.48550/arXiv.2201.13388","article_number":"2201.13388","date_created":"2023-08-22T14:23:16Z","year":"2022","department":[{"_id":"FrLo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2201.13388","open_access":"1"}],"oa_version":"Preprint","date_published":"2022-01-31T00:00:00Z","abstract":[{"lang":"eng","text":"Although reinforcement learning has seen remarkable progress over the last years, solving robust dexterous object-manipulation tasks in multi-object settings remains a challenge. In this paper, we focus on models that can learn manipulation tasks in fixed multi-object settings and extrapolate this skill zero-shot without any drop in performance when the number of objects changes. We consider the generic task of bringing a specific cube out of a set to a goal position. We find that previous approaches, which primarily leverage attention and graph neural network-based architectures, do not generalize their skills when the number of input objects changes while scaling as K2. We propose an alternative plug-and-play module based on relational inductive biases to overcome these limitations. Besides exceeding performances in their training environment, we show that our approach, which scales linearly in K, allows agents to extrapolate and generalize zero-shot to any new object number."}],"publication_status":"submitted","citation":{"ista":"Mambelli D, Träuble F, Bauer S, Schölkopf B, Locatello F. Compositional multi-object reinforcement learning with linear relation networks. arXiv, 2201.13388.","ieee":"D. Mambelli, F. Träuble, S. Bauer, B. Schölkopf, and F. Locatello, “Compositional multi-object reinforcement learning with linear relation networks,” arXiv. .","mla":"Mambelli, Davide, et al. “Compositional Multi-Object Reinforcement Learning with Linear Relation Networks.” ArXiv, 2201.13388, doi:10.48550/arXiv.2201.13388.","short":"D. Mambelli, F. Träuble, S. Bauer, B. Schölkopf, F. Locatello, ArXiv (n.d.).","chicago":"Mambelli, Davide, Frederik Träuble, Stefan Bauer, Bernhard Schölkopf, and Francesco Locatello. “Compositional Multi-Object Reinforcement Learning with Linear Relation Networks.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2201.13388.","ama":"Mambelli D, Träuble F, Bauer S, Schölkopf B, Locatello F. Compositional multi-object reinforcement learning with linear relation networks. arXiv. doi:10.48550/arXiv.2201.13388","apa":"Mambelli, D., Träuble, F., Bauer, S., Schölkopf, B., & Locatello, F. (n.d.). Compositional multi-object reinforcement learning with linear relation networks. arXiv. https://doi.org/10.48550/arXiv.2201.13388"},"arxiv":1,"_id":"14220"}]