{"id":590557,"date":"2019-06-01T14:14:40","date_gmt":"2019-06-01T21:14:40","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-event&p=590557"},"modified":"2019-07-08T08:55:51","modified_gmt":"2019-07-08T15:55:51","slug":"microsoft-icml-2019","status":"publish","type":"msr-event","link":"https:\/\/www.microsoft.com\/en-us\/research\/event\/microsoft-icml-2019\/","title":{"rendered":"Microsoft at ICML 2019"},"content":{"rendered":"

Venue:<\/strong> Long Beach Convention Center (opens in new tab)<\/span><\/a><\/p>\n

Website:<\/strong> ICLM 2019 (opens in new tab)<\/span><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

Venue: Long Beach Convention Center Website: ICLM 2019<\/p>\n","protected":false},"featured_media":590620,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_startdate":"2019-06-09","msr_enddate":"2019-06-15","msr_location":"Long Beach, California, USA","msr_expirationdate":"","msr_event_recording_link":"","msr_event_link":"https:\/\/icml.cc\/Register\/view-registration","msr_event_link_redirect":false,"msr_event_time":"","msr_hide_region":false,"msr_private_event":false,"footnotes":""},"research-area":[13556],"msr-region":[197900],"msr-event-type":[197941],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-590557","msr-event","type-msr-event","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-region-north-america","msr-event-type-conferences","msr-locale-en_us"],"msr_about":"Venue:<\/strong> Long Beach Convention Center<\/a>\r\n\r\nWebsite:<\/strong> ICLM 2019<\/a>","tab-content":[{"id":0,"name":"About","content":"Microsoft is excited to be a Gold sponsor of ICML<\/a>. We will have over 100 Microsoft attendees present at the conference. Stop by our booth (#310) to chat with our experts, see demos of our latest research and find out about career opportunities<\/a>\u00a0with Microsoft.\r\n

Microsoft Research attendees<\/h3>\r\nAdam Smiechowski\r\nAdith Swaminathan<\/a>\r\nAishwarya Rameshkumar\r\nAkshay Krishnamurthy\r\nAlekh Agarwal<\/a>\r\nAlessandro Sordoni<\/a>\r\nAlexey Taymanov<\/a>\r\nAmy Siebenthaler\r\nAndrew McNamara<\/a>\r\nAndrey Kolobov<\/a>\r\nAniket Anand Deshmukh\r\nBabak Aghazadeh\r\nBamdev Mishra\r\nBesmira Nushi<\/a>\r\nBhagirath Addepalli\r\nBianca Furtuna\r\nByron Changuion\r\nChandan Karadagur Ananda Reddy\r\nCharles Jacobs\r\nCheng Lu\r\nCheng Tan\r\nCheng Zhang\r\nchew-yean yam\r\nChicheng Zhang\r\nChris LaTerza\r\nChristian Borgs\r\nCrystal Schroeder\r\nDaniel Wilde\r\nDanny Garber\r\nDi He\r\nEhsan Vahedi\r\nForough Poursabzi-Sangdeh\r\nHal Daum\u00e9 III\r\nHarm van Seijen\r\nHonghao Qiu\r\nHuishuai Zhang\r\nIlya Razenshteyn\r\nJack Gerrits\r\nJacob Spoelstra\r\nJennifer Chayes\r\nJennifer Wortman Vaughan\r\nJohn Langford\r\nJustin Bronder\r\nKamil Ciosek\r\nKenneth Tran\r\nLester Mackey\r\nLevi Boyles\r\nLin Xiao\r\nLuca Saglietti\r\nLuke Stark\r\nMarkus Weimer\r\nMatineh Shaker\r\nMehdi Fatemi\r\nMihaela Curmei\r\nMirco Milletari\r\nMiroslav Dudik\r\nMiruna Oprescu\r\nNikhil Yadala\r\nNikos Karampatziakis\r\nOfer Dekel\r\nPatrick Lu\r\nPatty Ryan\r\nPaul Mineiro\r\nPengchuan Zhang\r\nPhilip Bachman\r\nPhilip Rosenfield\r\nPratik Kumar Jawanpuria\r\nPretesh Patel\r\nPriya Samnerkar\r\nPuneet Jolly\r\nQiuyuan Huang\r\nRan Gilad-Bachrach\r\nRemi Tachet des Combes\r\nREVANTH RAMESHKUMAR\r\nRich Caruana\r\nRobin McMahon\r\nRodrigo Kumpera\r\nRoland Fernandez\r\nRomain Laroche\r\nRuo-Chun Tzeng\r\nRyan Bae\r\nRyan Congdon\r\nRyota Tomioka\r\nSam Devlin\r\nSarah Bird\r\nSaurabh Bisht\r\nSebastian Kochman\r\nSebastian Tschiatschek\r\nSebastien Levy\r\nShauheen Zahirazami\r\nShize Su\r\nSoundararajan Srinivasan\r\nSrinagesh Sharma\r\nSteve Ballon\r\nSujeeth Bharadwaj\r\nSusan Dumais\r\nTao Qin\r\nTed Meeds\r\nTim Scarfe\r\nXiting Wang\r\nXu Tan\r\nXuan Zhao\r\nYihe Dong\r\nYingzhen Li\r\nYizhe Zhang\r\nZhe Wang"},{"id":1,"name":"Presentations","content":"

Microsoft presentation schedule<\/h2>\r\n

Sunday, June 9<\/h3>\r\n2:00 PM | Room 104 (Workshop)\r\nReal World Reinforcement Learning Workshop<\/strong>\r\nOrganizers: John Langford<\/strong><\/a>, Rodrigo Kumpera<\/strong>, Cheng Tan<\/strong><\/a>, Jack Gerrits<\/strong><\/a>, Paul Mineiro<\/strong><\/a>, Alexey Taymanov<\/strong><\/a>\r\n

Monday, June 10<\/h3>\r\n1:00 PM\u20133:15 PM | Grand Ballroom (Tutorial)\r\nNeural Approaches to Conversational AI<\/strong>\r\nMichel Galley<\/strong><\/a>, Jianfeng Gao<\/strong><\/a>\r\n

Tuesday, June 11<\/h3>\r\n11:25 AM\u201311:30 AM | Grand Ballroom (Oral)\r\nOn Certifying Non-Uniform Bounds against Adversarial Attacks<\/strong>\r\nChen Liu, Ryota Tomioka<\/strong><\/a>, Volkan Cevher\r\n\r\n11:40 AM\u201312:00 PM | Grand Ballroom (Oral)\r\nAdversarial Examples from Computational Constraints<\/strong>\r\nSebastien Bubeck<\/strong><\/a>, Yin Tat Lee, Eric Price, Ilya Razenshteyn<\/strong><\/a>\r\n\r\n11:40 AM\u201312:00 PM | Seaside Ballroom (Oral)\r\nContextual Memory Trees<\/strong>\r\nWen Sun, Alina Beygelzimer, Hal Daum\u00e9 III<\/strong><\/a>, John Langford<\/strong><\/a>, Paul Mineiro<\/strong><\/a>\r\n\r\n12:00 PM\u201312:05 PM | Room 104 (Oral)\r\nA Composite Randomized Incremental Gradient Method<\/strong>\r\nJunyu Zhang<\/strong>, Lin Xiao<\/strong><\/a>\r\n\r\n12:00 PM\u201312:05 PM | Room 101 (Oral)\r\nStein Point Markov Chain Monte Carlo<\/strong>\r\nWilson Ye Chen, Alessandro Barp, Francois-Xavier Briol, Jackson Gorham, Mark Girolami, Lester Mackey<\/strong><\/a>, Chris Oates\r\n\r\n2:00 PM\u20132:20 PM | Room 102 (Oral)\r\nGeneralized Approximate Survey Propagation for High-Dimensional Estimation<\/strong>\r\nCarlo Lucibello, Luca Saglietti<\/strong>, Yue Lu\r\n\r\n2:00 PM\u20132:20 PM | Grand Ballroom (Oral)\r\nOn Learning Invariant Representations for Domain Adaptation<\/strong>\r\nHan Zhao, Remi Tachet des Combes<\/strong><\/a>, Kun Zhang, Geoff Gordon<\/strong><\/a>\r\n\r\n2:00 PM\u20132:20 PM | Room 104 (Oral)\r\nSafe Policy Improvement with Baseline Bootstrapping<\/strong>\r\nRomain Laroche<\/strong><\/a>, Paul Trichelair, Remi Tachet des Combes<\/strong><\/a>\r\n\r\n4:40 PM\u20135:00 PM | Room 102 (Oral)\r\nLocally Private Bayesian Inference for Count Models<\/strong>\r\nAaron Schein, Zhiwei Steven Wu, Alexandra Schofield, Mingyuan Zhou, Hanna Wallach<\/strong><\/a>\r\n\r\n5:00 PM\u20135:05 PM | Room 102 (Oral)\r\nLow Latency Privacy Preserving Inference<\/strong>\r\nAlon Brutzkus<\/strong>, Ran Gilad-Bachrach<\/strong><\/a>, Oren Elisha<\/strong>\r\n\r\n5:10 PM\u20135:15 PM | Room 103 (Oral)\r\nProvably Efficient RL with Rich Observations via Latent State Decoding<\/strong>\r\nSimon Du, Akshay Krishnamurthy<\/strong><\/a>, Nan Jiang, Alekh Agarwal<\/strong><\/a>, Miroslav Dudik<\/strong>, John Langford<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #63)\r\nOn Certifying Non-Uniform Bounds against Adversarial Attacks<\/strong>\r\nChen Liu, Ryota Tomioka<\/strong><\/a>, Volkan Cevher\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #66)\r\nAdversarial Examples from Computational Constraints<\/strong>\r\nSebastien Bubeck<\/strong><\/a>, Yin Tat Lee, Eric Price, Ilya Razenshteyn<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #71)\r\nOn Learning Invariant Representations for Domain Adaptation<\/strong>\r\nHan Zhao, Remi Tachet des Combes<\/strong><\/a>, Kun Zhang, Geoff Gordon<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #97)\r\nA Composite Randomized Incremental Gradient Method<\/strong>\r\nJunyu Zhang<\/strong>, Lin Xiao<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #101)\r\nSafe Policy Improvement with Baseline Bootstrapping<\/strong>\r\nRomain Laroche<\/strong><\/a>, Paul Trichelair, Remi Tachet des Combes<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #125)\r\nContextual Memory Trees<\/strong>\r\nWen Sun, Alina Beygelzimer, Hal Daum\u00e9 III<\/strong><\/a>, John Langford<\/strong><\/a>, Paul Mineiro<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #160)\r\nGeneralized Approximate Survey Propagation for High-Dimensional Estimation<\/strong>\r\nCarlo Lucibello, Luca Saglietti<\/strong>, Yue Lu\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #175)\r\nLocally Private Bayesian Inference for Count Models<\/strong>\r\nAaron Schein, Zhiwei Steven Wu, Alexandra Schofield, Mingyuan Zhou, Hanna Wallach<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #176)\r\nLow Latency Privacy Preserving Inference<\/strong>\r\nAlon Brutzkus<\/strong>, Ran Gilad-Bachrach<\/strong><\/a>, Oren Elisha<\/strong>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #208)\r\nProvably Efficient RL with Rich Observations via Latent State Decoding<\/strong>\r\nSimon Du, Akshay Krishnamurthy<\/strong><\/a>, Nan Jiang, Alekh Agarwal<\/strong><\/a>, Miroslav Dudik<\/strong><\/a>, John Langford<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #216)\r\nStein Point Markov Chain Monte Carlo<\/strong>\r\nWilson Ye Chen, Alessandro Barp, Francois-Xavier Briol, Jackson Gorham, Mark Girolami, Lester Mackey<\/strong><\/a>, Chris Oates\r\n

Wednesday, June 12<\/h3>\r\n11:20 AM\u201311:25 AM | Room 201 (Oral)\r\nEfficient Amortised Bayesian Inference for Hierarchical and Nonlinear Dynamical Systems<\/strong>\r\nTed Meeds<\/strong><\/a>, Geoffrey Roeder, Paul Grant<\/strong><\/a>, Andrew Phillips<\/strong><\/a>, Neil Dalchau<\/strong><\/a>\r\n\r\n11:25 AM\u201311:30 AM | Hall A (Oral)\r\nAre Generative Classifiers More Robust to Adversarial Attacks?<\/strong>\r\nYingzhen Li<\/strong><\/a>, John Bradshaw, Yash Sharma\r\n\r\n11:40 AM\u201312:00 PM | Hall A (Oral)\r\nEDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAE<\/strong>\r\nChao Ma, Sebastian Tschiatschek<\/strong><\/a>, Konstantina Palla<\/strong><\/a>, Jose Hernandez-Lobato, Sebastian Nowozin, Cheng Zhang<\/strong><\/a>\r\n\r\n2:25 PM\u20132:30 PM | Room 201 (Oral)\r\nFast Context Adaptation via Meta-Learning<\/strong>\r\nLuisa Zintgraf, Kyriacos Shiarlis, Vitaly Kurin, Katja Hofmann<\/strong><\/a>, Shimon Whiteson\r\n\r\n2:35 PM\u20132:40 PM | Room 103 (Oral)\r\nOrthogonal Random Forest for Causal Inference<\/strong>\r\nMiruna Oprescu<\/strong><\/a>, Vasilis Syrgkanis<\/strong><\/a>, Zhiwei Steven Wu\r\n\r\n2:40 PM\u20133:00 PM | Room 101 (Oral)\r\nVariational Implicit Processes<\/strong>\r\nChao Ma, Yingzhen Li<\/strong><\/a>, Jose Hernandez-Lobato\r\n\r\n3:05 PM\u20133:10 PM | Room 104 (Oral)\r\nRiemannian adaptive stochastic gradient algorithms on matrix manifolds<\/strong>\r\nHiroyuki Kasai, Pratik Kumar Jawanpuria<\/strong>, Bamdev Mishra<\/strong>\r\n\r\n4:20 PM\u20134:25 PM | Grand Ballroom (Oral)\r\nAdaptive Neural Trees<\/strong>\r\nRyutaro Tanno, Kai Arulkumaran, Daniel Alexander, Antonio Criminisi<\/strong>, Aditya Nori<\/strong><\/a>\r\n\r\n4:20 PM\u20134:25 PM | Room 104 (Oral)\r\nDead-ends and Secure Exploration in Reinforcement Learning<\/strong>\r\nMehdi Fatemi<\/strong><\/a>, Shikhar Sharma<\/strong><\/a>, Harm van Seijen<\/strong><\/a>, Samira Ebrahimi Kahou<\/strong>\r\n\r\n4:20 PM\u20134:25 PM | Room 103 (Oral)\r\nSGD without Replacement: Sharper Rates for General Smooth Convex Functions<\/strong>\r\nDheeraj Nagaraj, Prateek Jain<\/strong><\/a>, Praneeth Netrapalli<\/strong><\/a>\r\n\r\n4:25 PM\u20134:30 PM | Grand Ballroom (Oral)\r\nConnectivity-Optimized Representation Learning via Persistent Homology<\/strong>\r\nChristoph Hofer, Roland Kwitt, Marc Niethammer, Mandar Dixit<\/strong>\r\n\r\n4:25 PM\u20134:30 PM | Room 201 (Oral)\r\nMyopic Posterior Sampling for Adaptive Goal Oriented Design of Experiments<\/strong>\r\nKirthevasan Kandasamy, Willie Neiswanger, Reed Zhang, Akshay Krishnamurthy<\/strong><\/a>, Jeff Schneider, Barnab\u00e1s P\u00f3czos\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #3)\r\nAre Generative Classifiers More Robust to Adversarial Attacks?<\/strong>\r\nYingzhen Li<\/strong><\/a>, John Bradshaw, Yash Sharma\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #6)\r\nEDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAE<\/strong>\r\nChao Ma, Sebastian Tschiatschek<\/strong><\/a>, Konstantina Palla<\/strong><\/a>, Jose Hernandez-Lobato, Sebastian Nowozin, Cheng Zhang<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #82)\r\nAdaptive Neural Trees<\/strong>\r\nRyutaro Tanno, Kai Arulkumaran, Daniel Alexander, Antonio Criminisi<\/strong>, Aditya Nori<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #83)\r\nConnectivity-Optimized Representation Learning via Persistent Homology<\/strong>\r\nChristoph Hofer, Roland Kwitt, Marc Niethammer, Mandar Dixit<\/strong>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #108)\r\nRiemannian adaptive stochastic gradient algorithms on matrix manifolds<\/strong>\r\nHiroyuki Kasai, Pratik Kumar Jawanpuria<\/strong>, Bamdev Mishra<\/strong>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #112)\r\nDead-ends and Secure Exploration in Reinforcement Learning<\/strong>\r\nMehdi Fatemi<\/strong><\/a>, Shikhar Sharma<\/strong><\/a>, Harm van Seijen<\/strong><\/a>, Samira Ebrahimi Kahou<\/strong>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #195)\r\nOrthogonal Random Forest for Causal Inference<\/strong>\r\nMiruna Oprescu<\/strong><\/a>, Vasilis Syrgkanis<\/strong><\/a>, Zhiwei Steven Wu\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #202)\r\nSGD without Replacement: Sharper Rates for General Smooth Convex Functions<\/strong>\r\nDheeraj Nagaraj, Prateek Jain<\/strong><\/a>, Praneeth Netrapalli<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #225)\r\nVariational Implicit Processes<\/strong>\r\nChao Ma, Yingzhen Li<\/strong><\/a>, Jose Hernandez-Lobato\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #241)\r\nEfficient Amortised Bayesian Inference for Hierarchical and Nonlinear Dynamical Systems<\/strong>\r\nTed Meeds<\/strong><\/a>, Geoffrey Roeder, Paul Grant<\/strong><\/a>, Andrew Phillips<\/strong><\/a>, Neil Dalchau<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #252)\r\nFast Context Adaptation via Meta-Learning<\/strong>\r\nLuisa Zintgraf, Kyriacos Shiarlis, Vitaly Kurin, Katja Hofmann<\/strong><\/a>, Shimon Whiteson\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #262)\r\nMyopic Posterior Sampling for Adaptive Goal Oriented Design of Experiments<\/strong>\r\nKirthevasan Kandasamy, Willie Neiswanger, Reed Zhang, Akshay Krishnamurthy<\/strong><\/a>, Jeff Schneider, Barnab\u00e1s P\u00f3czos\r\n

Thursday, June 13<\/h3>\r\n9:20 AM\u20139:25 AM | Grand Ballroom (Oral)\r\nTowards a Deep and Unified Understanding of Deep Neural Models in NLP<\/strong>\r\nChaoyu Guan, Xiting Wang<\/strong><\/a>, Quanshi Zhang, Runjin Chen, Di He<\/strong><\/a>, Xing Xie<\/strong><\/a>\r\n\r\n10:05 AM\u201310:10 AM | Room 201 (Oral)\r\nAlmost Unsupervised Text to Speech and Automatic Speech Recognition<\/strong>\r\nYi Ren, Xu Tan<\/strong><\/a>, Tao Qin<\/strong><\/a>, Sheng Zhao<\/strong>, Zhou Zhao, Tie-Yan Liu<\/strong><\/a>\r\n\r\n10:15 AM\u201310:30 AM | Room 102 (Oral)\r\nBandit Multiclass Linear Classification: Efficient Algorithms for the Separable Case<\/strong>\r\nAlina Beygelzimer, David Pal, Balazs Szorenyi, Devanathan Thiruvenkatachari, Chen-Yu Wei, Chicheng Zhang<\/strong><\/a>\r\n\r\n11:25 AM\u201311:30 AM | Room 102 (Oral)\r\nAdaptive Regret of Convex and Smooth Functions<\/strong>\r\nLijun Zhang, Tie-Yan Liu<\/strong><\/a>, Zhi-Hua Zhou\r\n\r\n11:30 AM\u201311:35 AM | Seaside Ballroom (Oral)\r\nFair Regression: Quantitative Definitions and Reduction-Based Algorithms<\/strong>\r\nAlekh Agarwal<\/strong><\/a>, Miroslav Dudik<\/strong><\/a>, Zhiwei Steven Wu\r\n\r\n11:35 AM\u201311:40 AM | Grand Ballroom (Oral)\r\nA Convergence Theory for Deep Learning via Over-Parameterization<\/strong>\r\nZeyuan Allen-Zhu<\/strong><\/a>, Yuanzhi Li, Zhao Song\r\n\r\n11:35 AM\u201311:40 AM | Hall B (Oral)\r\nNon-Monotonic Sequential Text Generation<\/strong>\r\nSean Welleck, Kiante Brantley, Hal Daum\u00e9 III<\/strong><\/a>, Kyunghyun Cho\r\n\r\n12:00 PM\u201312:05 PM | Room 104 (Oral)\r\nMASS: Masked Sequence to Sequence Pre-training for Language Generation<\/strong>\r\nKaitao Song, Xu Tan<\/strong><\/a>, Tao Qin<\/strong><\/a>, Jianfeng Lu, Tie-Yan Liu<\/strong><\/a>\r\n\r\n12:05 PM\u201312:10 PM | Room 104 (Oral)\r\nHumor in Word Embeddings: Cockamamie Gobbledegook for Nincompoops<\/strong>\r\nLimor Gultchin, Genevieve Patterson, Nancy Baym<\/strong><\/a>, Nathaniel Swinger, Adam Kalai<\/strong><\/a>\r\n\r\n12:15 PM\u201312:20 PM | Hall B (Oral)\r\nEfficient Training of BERT by Progressively Stacking<\/strong>\r\nLinyuan Gong, Di He<\/strong><\/a>, Zhuohan Li, Tao Qin<\/strong><\/a>, Liwei Wang, Tie-Yan Liu<\/strong><\/a>\r\n\r\n4:00 PM\u20134:20 PM | Hall B (Oral)\r\nDecentralized Exploration in Multi-Armed Bandits<\/strong>\r\nRaphael Feraud, Reda Alami, Romain Laroche<\/strong><\/a>\r\n\r\n4:20 PM\u20134:25 PM | Hall A (Oral)\r\nDistributed, Egocentric Representations of Graphs for Detecting Critical Structures<\/strong>\r\nRuo-Chun Tzeng<\/strong>, Shan-Hung (Brandon) Wu\r\n\r\n4:20 PM\u20134:25 PM | Hall B (Oral)\r\nWarm-starting Contextual Bandits: Robustly Combining Supervised and Bandit Feedback<\/strong>\r\nChicheng Zhang<\/strong><\/a>, Alekh Agarwal<\/strong><\/a>, Hal Daum\u00e9 III<\/strong><\/a>, John Langford<\/strong><\/a>, Sahand Negahban\r\n\r\n6:30 PM\u2013 9:00 PM | Pacific Ballroom (Poster #22)\r\nDistributed, Egocentric Representations of Graphs for Detecting Critical Structures<\/strong>\r\nRuo-Chun Tzeng<\/strong>, Shan-Hung (Brandon) Wu\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #45)\r\nNon-Monotonic Sequential Text Generation<\/strong>\r\nSean Welleck, Kiante Brantley, Hal Daum\u00e9 III<\/strong><\/a>, Kyunghyun Cho\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #50)\r\nEfficient Training of BERT by Progressively Stacking<\/strong>\r\nLinyuan Gong, Di He<\/strong><\/a>, Zhuohan Li, Tao Qin<\/strong><\/a>, Liwei Wang, Tie-Yan Liu<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #51)\r\nDecentralized Exploration in Multi-Armed Bandits<\/strong>\r\nRaphael Feraud, Reda Alami, Romain Laroche<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #52)\r\nWarm-starting Contextual Bandits: Robustly Combining Supervised and Bandit Feedback<\/strong>\r\nChicheng Zhang<\/strong><\/a>, Alekh Agarwal<\/strong><\/a>, Hal Daum\u00e9 III<\/a><\/strong>, John Langford<\/strong><\/a>, Sahand Negahban\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #62)\r\nTowards a Deep and Unified Understanding of Deep Neural Models in NLP<\/strong>\r\nChaoyu Guan, Xiting Wang<\/strong><\/a>, Quanshi Zhang, Runjin Chen, Di He<\/strong><\/a>, Xing Xie<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #75)\r\nA Convergence Theory for Deep Learning via Over-Parameterization<\/strong>\r\nZeyuan Allen-Zhu<\/strong><\/a>, Yuanzhi Li, Zhao Song\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #107)\r\nMASS: Masked Sequence to Sequence Pre-training for Language Generation<\/strong>\r\nKaitao Song, Xu Tan<\/strong><\/a>, Tao Qin<\/strong><\/a>, Jianfeng Lu, Tie-Yan Liu<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #108)\r\nHumor in Word Embeddings: Cockamamie Gobbledegook for Nincompoops<\/strong>\r\nLimor Gultchin, Genevieve Patterson, Nancy Baym<\/strong><\/a>, Nathaniel Swinger, Adam Kalai<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #132)\r\nFair Regression: Quantitative Definitions and Reduction-Based Algorithms<\/strong>\r\nAlekh Agarwal<\/strong><\/a>, Miroslav Dudik<\/strong><\/a>, Zhiwei Steven Wu\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #158)\r\nBandit Multiclass Linear Classification: Efficient Algorithms for the Separable Case<\/strong>\r\nAlina Beygelzimer, David Pal, Balazs Szorenyi, Devanathan Thiruvenkatachari, Chen-Yu Wei, Chicheng Zhang<\/strong><\/a>\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #161)\r\nAdaptive Regret of Convex and Smooth Functions<\/strong>\r\nLijun Zhang, Tie-Yan Liu<\/strong><\/a>, Zhi-Hua Zhou\r\n\r\n6:30 PM\u20139:00 PM | Pacific Ballroom (Poster #224)\r\nAlmost Unsupervised Text to Speech and Automatic Speech Recognition<\/strong>\r\nYi Ren, Xu Tan<\/strong><\/a>, Tao Qin<\/strong><\/a>, Sheng Zhao<\/strong>, Zhou Zhao, Tie-Yan Liu<\/strong><\/a>\r\n

Friday, June 14<\/h3>\r\n8:30 AM\u20136:00 PM | 104A (Workshop)\r\nClimate Change: How Can AI Help?<\/strong>\r\nCo-organizer: Jennifer Chayes<\/a>\r\n\r\n2:00 PM\u20136:00 PM | Seaside Ballroom (Workshop)\r\nReal-world Sequential Decision Making: Reinforcement Learning and Beyond<\/strong>\r\nCo-organizer: Adith Swaminathan<\/a>\r\n

Saturday, June 15<\/h3>\r\n8:30 AM\u20136:00 PM | 104B (Workshop)\r\nAI For Social Good<\/strong>\r\nCo-organizer: Lester Mackey<\/a>\r\n\r\n8:30 AM\u20136:00 PM | Seaside Ballroom (Workshop)\r\nAdaptive and Multitask Learning: Algorithms & Systems<\/strong>\r\nCo-organizer: Rich Caruana<\/a>\r\n\r\n8:30 AM\u20136:00 PM | 104A (Workshop)\r\nStein\u2019s Method for Machine Learning and Statistics<\/strong>\r\nCo-organizer: Lester Mackey<\/a>"}],"msr_startdate":"2019-06-09","msr_enddate":"2019-06-15","msr_event_time":"","msr_location":"Long Beach, California, USA","msr_event_link":"https:\/\/icml.cc\/Register\/view-registration","msr_event_recording_link":"","msr_startdate_formatted":"June 9, 2019","msr_register_text":"Watch now","msr_cta_link":"https:\/\/icml.cc\/Register\/view-registration","msr_cta_text":"Watch now","msr_cta_bi_name":"Event Register","featured_image_thumbnail":"\"Long","event_excerpt":"Microsoft is excited to be a Gold sponsor of ICML. We will have over 100 Microsoft attendees present at the conference. Stop by our booth (#310) to chat with our experts, see demos of our latest research and find out about career opportunities\u00a0with Microsoft. Microsoft Research attendees Adam Smiechowski Adith Swaminathan Aishwarya Rameshkumar Akshay Krishnamurthy Alekh Agarwal Alessandro Sordoni Alexey Taymanov Amy Siebenthaler Andrew McNamara Andrey Kolobov Aniket Anand Deshmukh Babak Aghazadeh Bamdev Mishra Besmira…","msr_research_lab":[],"related-researchers":[],"msr_impact_theme":[],"related-academic-programs":[],"related-groups":[],"related-projects":[],"related-opportunities":[],"related-publications":[580795,590737,592699,620094],"related-videos":[],"related-posts":[590116,590815,593650,609270],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/590557"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-event"}],"version-history":[{"count":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/590557\/revisions"}],"predecessor-version":[{"id":590563,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/590557\/revisions\/590563"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/590620"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=590557"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=590557"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=590557"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=590557"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=590557"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=590557"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=590557"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=590557"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=590557"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}