{"id":670011,"date":"2020-06-26T14:10:27","date_gmt":"2020-06-26T21:10:27","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-event&p=670011"},"modified":"2020-09-09T12:54:38","modified_gmt":"2020-09-09T19:54:38","slug":"icml-2020","status":"publish","type":"msr-event","link":"https:\/\/www.microsoft.com\/en-us\/research\/event\/icml-2020\/","title":{"rendered":"Microsoft at ICML 2020"},"content":{"rendered":"
Website:<\/strong> ICML 2020 (opens in new tab)<\/span><\/a><\/p>\n","protected":false},"excerpt":{"rendered":" Website: ICML 2020<\/p>\n","protected":false},"featured_media":392255,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_startdate":"2020-07-12","msr_enddate":"2020-07-18","msr_location":"Virtual\/Online","msr_expirationdate":"","msr_event_recording_link":"","msr_event_link":"","msr_event_link_redirect":false,"msr_event_time":"","msr_hide_region":false,"msr_private_event":false,"footnotes":""},"research-area":[13556],"msr-region":[256048,197900],"msr-event-type":[197941],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-670011","msr-event","type-msr-event","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-region-global","msr-region-north-america","msr-event-type-conferences","msr-locale-en_us"],"msr_about":"Website:<\/strong> ICML 2020<\/a>","tab-content":[{"id":0,"name":"About","content":"Microsoft is proud to be a Gold sponsor of the 37th International Conference on Machine Learning<\/a> (ICML), as well as Diamond sponsors at the 1st Women in Machine Learning Un-Workshop<\/a> and Platinum sponsors of the 4th Queer in AI Workshop<\/a>. We have over 50 papers accepted to the conference, and you can find details of our publications on the Accepted papers<\/a> and Workshops<\/a> tabs.\r\nCommittee chairs<\/h2>\r\nICML President: John Langford<\/a>\r\nICML Board Members: Hal Daum\u00e9 III<\/a>, Hanna Wallach<\/a>\r\nProgram Co-chair: Hal Daum\u00e9 III<\/a>\r\n
Invited speaker<\/h2>\r\n
Tuesday, July 14<\/h3>\r\n05:00 \u2013 06:45 PDT & 16:00 \u2013 17:45 PDT\r\nDoing Some Good with Machine Learning<\/strong>\r\nInvited Speaker: Lester Mackey<\/a>"},{"id":1,"name":"Sessions","content":"
Tuesday, July 14<\/h2>\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 20:00 \u2013 20:45 PDT\r\nNGBoost: Natural Gradient Boosting for Probabilistic Prediction<\/strong><\/a>\r\nTony Duan<\/strong>, Anand Avati, Daisy Ding, Khanh K. Thai, Sanjay Basu, Andrew Ng, Alejandro Schuler\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nOnline Learning for Active Cache Synchronization<\/strong><\/a>\r\nAndrey Kolobov<\/a>, Sebastien Bubeck<\/a>, Julian Zimmert\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nRandomized Smoothing of All Shapes and Sizes<\/strong><\/a>\r\nGreg Yang<\/a>, Tony Duan<\/strong>, J. Edward Hu<\/strong>, Hadi Salman<\/strong>, Ilya Razenshteyn<\/a>, Jerry Li<\/a>\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 20:00 \u2013 20:45 PDT\r\nPrivate Reinforcement Learning with PAC and Regret Guarantees<\/strong><\/a>\r\nGiuseppe Vietri, Borja de Balle Pigem, Akshay Krishnamurthy<\/a>, Steven Wu\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nScalable Nearest Neighbor Search for Optimal Transport<\/strong><\/a>\r\nArturs Backurs, Yihe Dong<\/strong>, Piotr Indyk, Ilya Razenshteyn<\/a>, Tal Wagner\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nCombinatorial Pure Exploration for Dueling Bandit<\/strong><\/a>\r\nWei Chen<\/a>, Yihan Du, Longbo Huang, Haoyu Zhao\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nDistance Metric Learning with Joint Representation Diversification<\/strong><\/a>\r\nXu Chu, Yang Lin, Xiting Wang<\/a>, Xin Gao, Qi Tong, Hailong Yu, Yasha Wang\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nEfficient Domain Generalization via Common-Specific Low-Rank Decomposition<\/strong><\/a>\r\nVihari Piratla<\/strong>, Praneeth Netrapalli<\/a>, Sunita Sarawagi\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nFaster Graph Embeddings via Coarsening<\/strong><\/a>\r\nMatthew Fahrbach, Gramoz Goranci, Sushant Sachdeva, Richard Peng, Chi Wang<\/a>\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nWhat is Local Optimality in Nonconvex-Nonconcave Minimax Optimization?<\/strong><\/a>\r\nChi Jin, Praneeth Netrapalli<\/a>, Michael Jordan\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nAn end-to-end approach for the verification problem: learning the right distance<\/strong><\/a>\r\nJoao Monteiro, Isabela Albuquerque, Jahangir Alam, R Devon Hjelm<\/a>, Tiago Falk\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nWorking Memory Graphs<\/strong><\/a>\r\nRicky Loynd<\/a>, Roland Fernandez<\/a>, Asli Celikyilmaz<\/a>, Adith Swaminathan<\/a>, Matthew Hausknecht<\/a>\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nInformative Dropout for Robust Representation Learning: A Shape-bias Perspective<\/strong><\/a>\r\nBaifeng Shi, Dinghuai Zhang, Qi Dai<\/a>, Jingdong Wang<\/a>, Zhanxing Zhu, Yadong Mu\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nNear-optimal Sample Complexity Bounds for Learning Latent\u00a0k\u2212polytopes and applications to Ad-Mixtures<\/strong><\/a>\r\nChiranjib Bhattacharyya, Ravindran Kannan<\/a>\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nDifferentially Private Set Union<\/strong><\/a>\r\nPankaj Gulhane<\/strong>, Sivakanth Gopi<\/a>, Janardhan Kulkarni<\/a>, Judy Hanwen Shen<\/strong>, Milad Shokouhi<\/a>, Sergey Yekhanin<\/a>\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nDiscount Factor as a Regularizer in Reinforcement Learning<\/strong><\/a>\r\nRon Amit, Kamil Ciosek<\/a>, Ron Meir\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nDROCC: Deep Robust One-Class Classification<\/strong><\/a>\r\nSachin Goyal<\/strong>, Aditi Raghunathan, Moksh Jain, Harsha Vardhan Simhadri<\/a>, Prateek Jain<\/a>\r\n\r\n09:00 \u2013 09:45 PDT\r\n2nd session: 20:00 \u2013 20:45 PDT\r\nFeature Quantization Improves GAN Training<\/strong><\/a>\r\nYang Zhao, Chunyuan Li<\/a>, Ping Yu, Jianfeng Gao<\/a>, Changyou Chen\r\n\r\n09:00 \u2013 09:45 PDT\r\n2nd session: 22:00 \u2013 22:45 PDT\r\nHow Good is the Bayes Posterior in Deep Neural Networks Really<\/strong><\/a>\r\nFlorian Wenzel, Kevin Roth, Bastiaan S. Veeling, Jakub Swiatkowski, Linh Tran, Stephan Mandt, Jasper Snoek, Tim Salimans, Rodolphe Jenatton, Sebastian Nowozin<\/a>\r\n\r\n11:00 \u2013 11:45 PDT\r\n2nd session: 22:00 \u2013 22:45 PDT\r\nOptimization and Analysis of the pAp@k Metric for Recommender Systems<\/strong><\/a>\r\nGaurush Hiranandani, Warut Vijitbenjaronk, Sanmi Koyejo, Prateek Jain<\/a>\r\n\r\n11:00 \u2013 11:45 PDT\r\n2nd session: 22:00 \u2013 22:45 PDT\r\nBandits with Adversarial Scaling<\/strong><\/a>\r\nThodoris Lykouris<\/a>, Vahab Mirrokni, Renato Leme\r\n\r\n12:00 \u2013 12:45 PDT\r\n2nd session: July 15 | 01:00 \u2013 01:45 PDT\r\nTaskNorm: Rethinking Batch Normalization for Meta-Learning<\/strong><\/a>\r\nJohn Bronskill, Jonathan Gordon, James Requeima, Sebastian Nowozin<\/a>, Richard E. Turner\r\n\r\n13:00 \u2013 13:45 PDT\r\n2nd session: July 15 | 01:00 \u2013 01:45 PDT\r\nGNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulation<\/strong><\/a>\r\nMarc Brockschmidt<\/a>\r\n\r\n18:00 \u2013 18:45 PDT\r\nOnline Learning for Active Cache Synchronization<\/strong><\/a>\r\nAndrey Kolobov<\/a>, Sebastien Bubeck<\/a>, Julian Zimmert\r\n\r\n18:00 \u2013 18:45 PDT\r\nScalable Nearest Neighbor Search for Optimal Transport<\/strong><\/a>\r\nArturs Backurs, Yihe Dong<\/strong>, Piotr Indyk, Ilya Razenshteyn<\/a>, Tal Wagner\r\n\r\n18:00 \u2013 18:45 PDT\r\nFaster Graph Embeddings via Coarsening<\/strong><\/a>\r\nMatthew Fahrbach, Gramoz Goranci, Sushant Sachdeva, Richard Peng, Chi Wang<\/a>\r\n\r\n18:00 \u2013 18:45 PDT\r\nWhat is Local Optimality in Nonconvex-Nonconcave Minimax Optimization?<\/strong><\/a>\r\nChi Jin, Praneeth Netrapalli<\/a>, Michael Jordan\r\n\r\n19:00 \u2013 19:45 PDT\r\nRandomized Smoothing of All Shapes and Sizes<\/strong><\/a>\r\nGreg Yang<\/a>, Tony Duan<\/strong>, J. Edward Hu<\/strong>, Hadi Salman<\/strong>, Ilya Razenshteyn<\/a>, Jerry Li<\/a>\r\n\r\n19:00 \u2013 19:45 PDT\r\nAn end-to-end approach for the verification problem: learning the right distance<\/strong><\/a>\r\nJoao Monteiro, Isabela Albuquerque, Jahangir Alam, R Devon Hjelm<\/a>, Tiago Falk\r\n\r\n19:00 \u2013 19:45 PDT\r\nCombinatorial Pure Exploration for Dueling Bandit<\/strong><\/a>\r\nWei Chen<\/a>, Yihan Du, Longbo Huang, Haoyu Zhao\r\n\r\n19:00 \u2013 19:45 PDT\r\nDistance Metric Learning with Joint Representation Diversification<\/strong><\/a>\r\nXu Chu, Yang Lin, Xiting Wang<\/a>, Xin Gao, Qi Tong, Hailong Yu, Yasha Wang\r\n\r\n19:00 \u2013 19:45 PDT\r\nEfficient Domain Generalization via Common-Specific Low-Rank Decomposition<\/strong><\/a>\r\nVihari Piratla<\/strong>, Praneeth Netrapalli<\/a>, Sunita Sarawagi\r\n\r\n19:00 \u2013 19:45 PDT\r\nInformative Dropout for Robust Representation Learning: A Shape-bias Perspective<\/strong><\/a>\r\nBaifeng Shi, Dinghuai Zhang, Qi Dai<\/a>, Jingdong Wang<\/a>, Zhanxing Zhu, Yadong Mu\r\n\r\n19:00 \u2013 19:45 PDT\r\nDifferentially Private Set Union<\/strong><\/a>\r\nPankaj Gulhane<\/strong>, Sivakanth Gopi<\/a>, Janardhan Kulkarni<\/a>, Judy Hanwen Shen<\/strong>, Milad Shokouhi<\/a>, Sergey Yekhanin<\/a>\r\n\r\n20:00 \u2013 20:45 PDT\r\nNGBoost: Natural Gradient Boosting for Probabilistic Prediction<\/strong><\/a>\r\nTony Duan<\/strong>, Anand Avati, Daisy Ding, Khanh K. Thai, Sanjay Basu, Andrew Ng, Alejandro Schuler\r\n\r\n20:00 \u2013 20:45 PDT\r\nPrivate Reinforcement Learning with PAC and Regret Guarantees<\/strong><\/a>\r\nGiuseppe Vietri, Borja de Balle Pigem, Akshay Krishnamurthy<\/a>, Steven Wu\r\n\r\n20:00 \u2013 20:45 PDT\r\nFeature Quantization Improves GAN Training<\/strong><\/a>\r\nYang Zhao, Chunyuan Li<\/a>, Ping Yu, Jianfeng Gao<\/a>, Changyou Chen\r\n\r\n21:00 \u2013 21:45 PDT\r\nWorking Memory Graphs<\/strong><\/a>\r\nRicky Loynd<\/a>, Roland Fernandez<\/a>, Asli Celikyilmaz<\/a>, Adith Swaminathan<\/a>, Matthew Hausknecht<\/a>\r\n\r\n21:00 \u2013 21:45 PDT\r\nNear-optimal Sample Complexity Bounds for Learning Latent\u00a0k\u2212polytopes and applications to Ad-Mixtures<\/strong><\/a>\r\nChiranjib Bhattacharyya, Ravindran Kannan<\/a>\r\n\r\n21:00 \u2013 21:45 PDT\r\nDiscount Factor as a Regularizer in Reinforcement Learning<\/strong><\/a>\r\nRon Amit, Kamil Ciosek<\/a>, Ron Meir\r\n\r\n21:00 \u2013 21:45 PDT\r\nDROCC: Deep Robust One-Class Classification<\/strong><\/a>\r\nSachin Goyal<\/strong>, Aditi Raghunathan, Moksh Jain, Harsha Vardhan Simhadri<\/a>, Prateek Jain<\/a>\r\n\r\n22:00 \u2013 22:45 PDT\r\nOptimization and Analysis of the pAp@k Metric for Recommender Systems<\/strong><\/a>\r\nGaurush Hiranandani, Warut Vijitbenjaronk, Sanmi Koyejo, Prateek Jain<\/a>\r\n\r\n22:00 \u2013 22:45 PDT\r\nBandits with Adversarial Scaling<\/strong><\/a>\r\nThodoris Lykouris<\/a>, Vahab Mirrokni, Renato Leme\r\n\r\n22:00 \u2013 22:45 PDT\r\nHow Good is the Bayes Posterior in Deep Neural Networks Really<\/strong><\/a>\r\nFlorian Wenzel, Kevin Roth, Bastiaan S. Veeling, Jakub Swiatkowski, Linh Tran, Stephan Mandt, Jasper Snoek, Tim Salimans, Rodolphe Jenatton, Sebastian Nowozin<\/a>\r\n\r\n
\r\n\r\nWednesday, July 15<\/h2>\r\n01:00 \u2013 01:45 PDT\r\nGNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulation<\/strong><\/a>\r\nMarc Brockschmidt<\/a>\r\n\r\n01:00 \u2013 01:45 PDT\r\nTaskNorm: Rethinking Batch Normalization for Meta-Learning<\/strong><\/a>\r\nJohn Bronskill, Jonathan Gordon, James Requeima, Sebastian Nowozin<\/a>, Richard E. Turner\r\n\r\n05:00 \u2013 05:45 PDT\r\n2nd session: 16:00 \u2013 16:45 PDT\r\nAdaptive Estimator Selection for Off-Policy Evaluation<\/strong><\/a>\r\nYi Su, Pavithra Srinath<\/a>, Akshay Krishnamurthy<\/a>\r\n\r\n05:00 \u2013 05:45 PDT\r\n2nd session: 16:00 \u2013 16:45 PDT\r\nPrivately Learning Markov Random Fields<\/strong><\/a>\r\nGautam Kamath, Janardhan Kulkarni<\/a>, Steven Wu, Huanyu Zhang\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nThe Non-IID Data Quagmire of Decentralized Machine Learning<\/strong><\/a>\r\nKevin Hsieh<\/strong>, Amar Phanishayee<\/a>, Onur Mutlu, Phillip Gibbons\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nAlleviating Privacy Attacks via Causal Learning<\/strong><\/a>\r\nShruti Tople<\/a>, Amit Sharma<\/a>, Aditya Nori<\/a>\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\n(Locally) Differentially Private Combinatorial Semi-Bandits<\/strong><\/a>\r\nXiaoyu Chen, Kai Zheng, Zixin Zhou, Yunchang Yang, Wei Chen<\/a>, Liwei Wang\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 20:00 \u2013 20:45 PDT\r\nThe Usual Suspects? Reassessing Blame for VAE Posterior Collapse<\/strong><\/a>\r\nBin Dai, Ziyu Wang, David Wipf<\/strong>\r\n\r\n10:00 \u2013 10:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nSingle Point Transductive Prediction<\/strong><\/a>\r\nNilesh Tripuraneni, Lester Mackey<\/a>\r\n\r\n10:00 \u2013 10:45 PDT\r\n2nd session: 21:00 \u2013 21:45 PDT\r\nLearning Calibratable Policies using Programmatic Style-Consistency<\/strong><\/a>\r\nEric Zhan, Albert Tseng, Yisong Yue, Adith Swaminathan<\/a>, Matthew Hausknecht<\/a>\r\n\r\n11:00 \u2013 11:45 PDT\r\n2nd session: 22:00 \u2013 22:45 PDT\r\nStatistically Preconditioned Accelerated Gradient Method for Distributed Optimization<\/strong><\/a>\r\nHadrien Hendrikx, Lin Xiao<\/a>, Sebastien Bubeck<\/a>, Francis Bach, Laurent Massouli\u00e9<\/strong>\r\n\r\n12:00 \u2013 12:45 PDT\r\n2nd session: July 16 | 01:00 \u2013 01:45 PDT\r\nNeuro-Symbolic Visual Reasoning: Disentangling \"Visual\" from \"Reasoning\"<\/strong><\/a>\r\nSaeed Amizadeh<\/strong>, Hamid Palangi<\/a>, Oleksandr Polozov<\/a>, Yichen Huang<\/strong>, Kazuhito Koishida<\/a>\r\n\r\n16:00 \u2013 16:45 PDT\r\n2nd session: July 16 | 03:00 \u2013 03:45 PDT\r\nOptimization from Structured Samples for Coverage Functions<\/strong><\/a>\r\nWei Chen<\/a>, Xiaoming Sun, Jialin Zhang, Zhijie Zhang\r\n\r\n16:00 \u2013 16:45 PDT\r\nAdaptive Estimator Selection for Off-Policy Evaluation<\/strong><\/a>\r\nYi Su, Pavithra Srinath<\/a>, Akshay Krishnamurthy<\/a>\r\n\r\n16:00 \u2013 16:45 PDT\r\nPrivately Learning Markov Random Fields<\/strong><\/a>\r\nGautam Kamath, Janardhan Kulkarni<\/a>, Steven Wu, Huanyu Zhang\r\n\r\n20:00 \u2013 20:45 PDT\r\nThe Usual Suspects? Reassessing Blame for VAE Posterior Collapse<\/strong><\/a>\r\nBin Dai, Ziyu Wang, David Wipf<\/strong>\r\n\r\n21:00 \u2013 21:45 PDT\r\nThe Non-IID Data Quagmire of Decentralized Machine Learning<\/strong><\/a>\r\nKevin Hsieh<\/strong>, Amar Phanishayee<\/a>, Onur Mutlu, Phillip Gibbons\r\n\r\n21:00 \u2013 21:45 PDT\r\nSingle Point Transductive Prediction<\/strong><\/a>\r\nNilesh Tripuraneni, Lester Mackey<\/a>\r\n\r\n21:00 \u2013 21:45 PDT\r\nAlleviating Privacy Attacks via Causal Learning<\/strong><\/a>\r\nShruti Tople<\/a>, Amit Sharma<\/a>, Aditya Nori<\/a>\r\n\r\n21:00 \u2013 21:45 PDT\r\n(Locally) Differentially Private Combinatorial Semi-Bandits<\/strong><\/a>\r\nXiaoyu Chen, Kai Zheng, Zixin Zhou, Yunchang Yang, Wei Chen<\/a>, Liwei Wang\r\n\r\n21:00 \u2013 21:45 PDT\r\nLearning Calibratable Policies using Programmatic Style-Consistency<\/strong><\/a>\r\nEric Zhan, Albert Tseng, Yisong Yue, Adith Swaminathan<\/a>, Matthew Hausknecht<\/a>\r\n\r\n22:00 \u2013 22:45 PDT\r\nStatistically Preconditioned Accelerated Gradient Method for Distributed Optimization<\/strong><\/a>\r\nHadrien Hendrikx, Lin Xiao<\/a>, Sebastien Bubeck<\/a>, Francis Bach, Laurent Massouli\u00e9<\/strong>\r\n\r\n
\r\n\r\nThursday, July 16<\/h2>\r\n01:00 \u2013 01:45 PDT\r\nNeuro-Symbolic Visual Reasoning: Disentangling \"Visual\" from \"Reasoning\"<\/strong><\/a>\r\nSaeed Amizadeh<\/strong>, Hamid Palangi<\/a>, Oleksandr Polozov<\/a>, Yichen Huang<\/strong>, Kazuhito Koishida<\/a>\r\n\r\n03:00 \u2013 03:45 PDT\r\nOptimization from Structured Samples for Coverage Functions<\/strong><\/a>\r\nWei Chen<\/a>, Xiaoming Sun, Jialin Zhang, Zhijie Zhang\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: July 17 | 18:00 \u2013 18:45 PDT\r\nMapping Natural-language Problems to Formal-language Solutions Using Structured Neural Representations<\/strong><\/a>\r\nKezhen Chen, Qiuyuan Huang<\/a>, Hamid Palangi<\/a>, Paul Smolensky<\/a>, Ken Forbus, Jianfeng Gao<\/a>\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 17:00 \u2013 17:45 PDT\r\nBINOCULARS for efficient, nonmyopic sequential experimental design<\/strong><\/a>\r\nShali Jiang, Henry Chai, Javier Gonzalez<\/a>, Roman Garnett\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nBlack-Box Methods for Restoring Monotonicity<\/strong><\/a>\r\nEvangelia Gergatsouli, Brendan Lucier<\/a>, Christos Tzamos\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 17:00 \u2013 17:45 PDT\r\nCLUB: A Contrastive Log-ratio Upper Bound of Mutual Information<\/strong><\/a>\r\nPengyu Cheng, Weituo Hao, Shuyang Dai, Jiachang Liu, Zhe Gan<\/strong>, Lawrence Carin\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 17:00 \u2013 17:45 PDT\r\nNeural Datalog Through Time: Informed Temporal Modeling via Logical Specification<\/strong><\/a>\r\nHongyuan Mei, Guanghui Qin, Minjie Xu, Jason Eisner<\/strong>\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nProvably Efficient Model-based Policy Adaptation<\/strong><\/a>\r\nYuda Song, Aditi Mavalankar, Wen Sun<\/strong>, Sicun Gao\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nSequence Generation with Mixed Representations<\/strong><\/a>\r\nLijun Wu<\/strong>, Shufang Xie<\/strong>, Yingce Xia, Yang Fan, Jian-Huang Lai, Tao Qin<\/a>, Tie-Yan Liu<\/a>\r\n\r\n06:00 \u2013 06:45 PDT\r\n2nd session: 17:00 \u2013 17:45 PDT\r\nReward-Free Exploration for Reinforcement Learning<\/strong><\/a>\r\nChi Jin, Akshay Krishnamurthy<\/a>, Max Simchowitz, Tiancheng Yu\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nNo-Regret and Incentive-Compatible Online Learning<\/strong><\/a>\r\nRupert Freeman<\/a>, David Pennock, Charikleia Podimata, Jennifer Wortman Vaughan<\/a>\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nGraph Optimal Transport for Cross-Domain Alignment<\/strong><\/a>\r\nLiqun Chen, Zhe Gan<\/strong>, Yu Cheng<\/strong>, Linjie Li<\/strong>, Lawrence Carin, Jingjing Liu<\/a>\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 20:00 \u2013 20:45 PDT\r\nDoubly Robust Off-policy Evaluation with Shrinkage<\/strong><\/a>\r\nYi Su, Maria Dimakopoulou, Akshay Krishnamurthy<\/a>, Miroslav Dudik<\/a>\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 18:00 \u2013 18:45 PDT\r\nVariance Reduction and Quasi-Newton for Particle-Based Variational Inference<\/strong><\/a> Michael Zhu, Chang Liu<\/strong>, Jun Zhu\r\n\r\n07:00 \u2013 07:45 PDT\r\n2nd session: 20:00 \u2013 20:45 PDT\r\nKinematic State Abstraction and Provably Efficient Rich-Observation Reinforcement Learning<\/strong><\/a>\r\nDipendra Misra<\/a>, Mikael Henaff<\/a>, Akshay Krishnamurthy<\/a>, John Langford<\/a>\r\n\r\n08:00 \u2013 08:45 PDT\r\n2nd session: 19:00 \u2013 19:45 PDT\r\nUniLMv2: Pseudo-Masked Language Models for Unified Language Model Pre-Training<\/strong><\/a>\r\nHangbo Bao, Li Dong<\/a>, Furu Wei<\/a>, Wenhui Wang<\/strong>, Nan Yang<\/a>, Xiaodong Liu<\/a>, Yu Wang<\/strong>, Jianfeng Gao<\/a>, Songhao Piao, Ming Zhou<\/a>, Hsiao-Wuen Hon<\/a>\r\n\r\n09:00 \u2013 09:45 PDT\r\n2nd session: 23:00 \u2013 23:45 PDT\r\nBounding the fairness and accuracy of classifiers from population statistics<\/strong><\/a>\r\nSivan Sabato<\/strong>, Elad Yom-Tov<\/a>\r\n\r\n12:00 \u2013 12:45 PDT\r\n2nd session: July 17 | 00:00 \u2013 00:45 PDT\r\nSoft Threshold Weight Reparameterization for Learnable Sparsity<\/strong><\/a>\r\nAditya Kusupati, Vivek Ramanujan, Raghav Somani, Mitchell Wortsman, Prateek Jain<\/a>, Sham Kakade, Ali Farhadi\r\n\r\n12:00 \u2013 12:45 PDT\r\n2nd session: July 17 | 01:00 \u2013 01:45 PDT\r\nThe k-tied Normal Distribution: A Compact Parameterization of Gaussian Mean Field Posteriors in Bayesian Neural Networks<\/strong><\/a>\r\nJakub Swiatkowski, Kevin Roth, Bastiaan S. Veeling, Linh Tran, Joshua V. Dillon, Stephan Mandt, Jasper Snoek, Tim Salimans, Rodolphe Jenatton, Sebastian Nowozin<\/a>\r\n\r\n17:00 \u2013 17:45 PDT\r\nBINOCULARS for efficient, nonmyopic sequential experimental design<\/strong><\/a>\r\nShali Jiang, Henry Chai, Javier Gonzalez<\/a>, Roman Garnett\r\n\r\n17:00 \u2013 17:45 PDT\r\nCLUB: A Contrastive Log-ratio Upper Bound of Mutual Information<\/strong><\/a>\r\nPengyu Cheng, Weituo Hao, Shuyang Dai, Jiachang Liu, Zhe Gan<\/strong>, Lawrence Carin\r\n\r\n17:00 \u2013 17:45 PDT\r\nNeural Datalog Through Time: Informed Temporal Modeling via Logical Specification<\/strong><\/a>\r\nHongyuan Mei, Guanghui Qin, Minjie Xu, Jason Eisner<\/strong>\r\n\r\n17:00 \u2013 17:45 PDT\r\nReward-Free Exploration for Reinforcement Learning<\/strong><\/a>\r\nChi Jin, Akshay Krishnamurthy<\/a>, Max Simchowitz, Tiancheng Yu\r\n\r\n18:00 \u2013 18:45 PDT\r\nMapping Natural-language Problems to Formal-language Solutions Using Structured Neural Representations<\/strong><\/a>\r\nKezhen Chen, Qiuyuan Huang<\/a>, Hamid Palangi<\/a>, Paul Smolensky<\/a>, Ken Forbus, Jianfeng Gao<\/a>\r\n\r\n18:00 \u2013 18:45 PDT\r\nNo-Regret and Incentive-Compatible Online Learning<\/strong><\/a>\r\nRupert Freeman<\/a>, David Pennock, Charikleia Podimata, Jennifer Wortman Vaughan<\/a>\r\n\r\n18:00 \u2013 18:45 PDT\r\n2nd session: July 17 | 04:00 \u2013 04:45 PDT\r\nOn Layer Normalization in the Transformer Architecture<\/strong><\/a>\r\nRuibin Xiong, Yunchang Yang, Di He, Kai Zheng, Shuxin Zheng<\/a>, Chen Xing, Huishuai Zhang<\/a>, Yanyan Lan, Liwei Wang, Tie-Yan Liu<\/a>\r\n\r\n18:00 \u2013 18:45 PDT\r\nBlack-Box Methods for Restoring Monotonicity<\/strong><\/a>\r\nEvangelia Gergatsouli,