Microsoft at NeurIPS 2019

Microsoft at NeurIPS 2019

About

Microsoft is a proud Diamond sponsor of the 33rd annual conference on Neural Information Processing Systems. Over 300 of our researchers are involved in spotlight sessions, presentations, symposiums, posters, accepted papers, and workshops. Stop by our booth (#301) to chat with our experts, see demos of our latest research and find out about career opportunities with Microsoft.

Organizing Committee

Hanna Wallach, General Chair
Danielle Belgrave, Tutorial Chair
Jenn Wortman Vaughan, Workshop Chair

Neural Information Processing Systems Foundation Board

Hanna Wallach, Board member

Microsoft Attendees

Ashkan Aazami
Eslam Abdelreheem
Shady Abdelrehim
Mohamed Abedelmalik
Robin Abraham
Alekh Agarwal
Saurabh Agarwal
Mehdi Aghagolzadeh
John Agosta
Robert Aichner
Pelkins Ajanoh
Zeyuan Allen-Zhu
Javier Alvarez-Valle
Anton Amirov
Emmanuel Awa
Ahmed Awadallah
Philip Bachman
Niranjan Baddam
Solon Barocas
Emad Barsoum
Danielle Belgrave
Paul Bennett
Ebrahim Beyrami
Sujeeth Bharadwaj
Anuj Bhatia
Enes Bilgin
Sarah Bird
Christopher Bishop
Christian Borgs
Chris Brockett
Marc Brockschmidt
Justin Bronder
Sebastien Bubeck
William Buchwalter
James Budnik
Alejandro Buendia
Guihong Cao
Rich Caruana
Asli Celikyilmaz
Urszula Chajewska
Doran Chakraborty
Preetom Chakraborty
Wang Chao
Jennifer Chayes
Deqing Chen
Wei Chen
Isabel Chien
Kamil Ciosek
Ryan Congdon
Ana-Maria Constantin
Marc-Alexandre Côté
Camille Couturier
Markus Cozowicz
Daniel Crankshaw
Carlo Curino
Ross Cutler
Bita Darvish Rouhani
Hal Daumé III
Dave DeBarr
Ofer Dekel
Crystal Deng
Sarah D’Ettorre
Sam Devlin
Debadeepta Dey
Gabriel Dominguez Conde
Haoyu Dong
Li Dong
Yihe Dong
Nan Duan
Miro Dudik
Susan Dumais
Jason Eisner
Joyce Fang
Mehdi Fatemi
Tao Feng
Roland Fernandez
Mikael Figueroa
Senja Filipi
Nicole Fitzgerald
Andrew Fitzgibbon
Rana Forsati
Victor Fragoso
Nicolo Fusi
Michel Galley
Zhe Gan
Kris Ganjam
Xiang Gao
Raluca Georgescu
Ran Gilad-Bachrach
Igor Gitman
Garrett Goh
Vikas Gosain
Suriya Gunasekar
Devin Gunson
Xuenan Guo
Abhishek Gupta
Nicholas Gyde
Benjamin Han
Judy Hanwen Shen
Matthew Hausknecht
Xiang He
Yuxiong He
Allison Hegel
Maggie Hei
Mikael Henaff
R Devon Hjelm
Katja Hofmann
Eric Horvitz
Saghar Hosseini
Jason Hsu
Shih-Chung Hsu
Edward Hu
Yameng Huang
Stephanie Hyland
Scott Inglis
Matteo Interlandi
Prateek Jain
Shantanu Jain
Mihai Jalobeanu
Minwoo Jeong
Henry Jerez
Smriti Jha
Anup Kadkol
Adam Kalai
Keiji Kanazawa
Panashe Kanengoni
Ashish Kapoor
Nikos Karampatziakis
Meha Kaushik
Emre Kiciman
Aerin Kim
Julia Kiseleva
Andrey Kolobov
XC Kong
Harish Krishna
Akshay Krishnamurthy
Sandip Kulkarni
Samir Kumar
Noboru Kuno
Nate Kushman
MinKyoung Kang
Hunter Lang
John Langford
Michael Lazos
Xuan Li
Yi Li
Yingzhen Li
Zhen Li
Zhuowei Li
Olga Liakhovich
Hongwei Liang
Fang Liu
Tie-Yan Liu
Daniel Lo
Cheng Lu
Brendan Lucier
Roman Lutz
Patrick MacAlpine
Lester Mackey
Ratnesh Madaan
Divyat Mahajan
Journey McDowell
Daniel McDuff
Andrew McNamara
Ted Meeds
Christopher Meek
Soroush Mehri
Levi Melnick
Vanessa Milan
Paul Mineiro
Anna Mitenkova
Azadeh Mobasher
Caitlin Mullock
Tsendsuren Munkhdalai
Cameron Musco
Mridu Narang
Tristan Naumann
Praneeth Netrapalli
Fanny Nina Paravecino
Elnaz Noori
Harsha Nori
Ehi Nosakhare
Besmira Nushi
Tim O’Brien
Cassandra Oduola
Adrian O’Grady
Oreoluwa Ogundipe
Olga Ohrimenko
Ozan Oktay
Miruna Oprescu
Hamid Palangi
Praveen Palanisamy
Ravi Pandya
Kaushal Paneri
Abhishek Panigrahi
Mikhail Parakhin
Raj Parihar
Devangkumar Patel
Vanja Paunic
Andi Peng
Dmitrij Petters
Vasilii Pisar
Alex Polozov
Hoifung Poon
Peter Potash
Forough Poursabzi-Sangdeh
Ishita Prasad
Tao Qin
Chris Quirk
Sravanthi Rajanala
Saravanakumar Rajmohan
Gonzalo Ramos
Abhishek Rao
Ilya Razenshteyn
Michael Revow
Jordi Ribas
Matt Richardson
Tyler Romero
Nathaniel Rose
Corbin Rosset
Marco Rossi
Victor Ruehle
Patricia Ryan
Hadi Salman
Mathew Salvaris
Tobias Schnabel
Crystal Schroeder
Siddhartha Sen
Samira Shabanian
Matineh Shaker
Richika Sharan
Amit Sharma
Vighnesh Shiv
Victor Shnayder
Yuanchao Shu
Amy Siebenthaler
Jonathan Sinai
Aleksandrs Slivkins
Bryan Smith
Brandon Smock
Paul Smolensky
Yale Song
Alessandro Sordoni
Anirudh Srinivasan
Yann Stadnicki
Luke Stark
William Stasior
Ross Story
Paul Stubbs
Jan Stuehmer
Shih-Chieh Su
Shize Su
Kaheer Suleman
Nick Switanek
Vasilis Syrgkanis
Saurabh Tiwary
Remi Tachet des Combes
Abe Taha
Xu Tan
Yi Tang
Ryutaro Tanno
Ryota Tomioka
Kenneth Tran
Marc Tremblay
Anusua Trivedi
Sebastian Tschiatschek
Rutger van Haasteren
Tempest van Schaik
Harm Van Seijen
Harsha Vardhan Simhadri
Sai Vemprala
Gangadharan Venkatasubramanian
Govert Verkes
Evelyne Viegas
Felipe Vieira Frujeri
Vibhav Vineet
Duncan Wadsworth
Chi Wang
Jingyan Wang
Junhua Wang
Markus Weimer
Nile Wilson
David Wipf
Jennifer Wortman Vaughan
Lin Xiao
Faith Xu
Yixi Xu
Xinwei Xue
Yadollah Yaghoobzadeh
Yulan Yan
Greg Yang
Siyu Yang
Sergey Yekhanin
Legg Yeung
Ali Zaidi
Yordan Zaykov
Cheng Zhang
Chicheng Zhang
Dongmei Zhang
Hang Zhang
Haozhe Zhang
Huishuai Zhang
Jack Zhang
Weixing Zhang
Yizhe Zhang
Jing Zhao
Li Zhao
Lingling Zheng
Zhiyong Zheng
Chenguang Zhu

Tutorials + Workshops

Tutorials

Monday, December 9

8:30 AM–10:30 AM | West Hall C + B3
Imitation Learning and its Application to Natural Language Generation
Kyunghyun Cho, Hal Daumé III

2:45 PM–4:45 PM | West Hall C + B3
Reinforcement Learning: Past, Present, and Future Perspectives
Katja Hofmann

Workshops

Sunday, December 8

9:00 AM–1:00 PM | West 208-209
Vowpal Wabbit for real world reinforcement learning
Organizers: John Langford, Rodrigo Kumpera, Jack Gerrits, Yann Stadnicki, Marco Rossi

Monday, December 9

2:00 PM–5:00 PM | East 8 + 15
Queer in AI
Co-organizers: Andrew McNamara, Luke Stark
Panelist: Hanna Wallach

6:30 PM–8:00 PM | East Exhibition Hall B | Joint Affinity Groups Poster Session

Co-opNet: Cooperative Generator-Discriminator Networks for Abstractive Summarization with Narrative Flow
Saadia Gabriel, Antoine Bosselut, Ari Holtzman, Jan Buys, Kyle Lo, Asli Celikyilmaz, Yejin Choi

Non-Monotonic Sequential Text Generation
Kianté Brantley, Hal Daumé III, Kyunghyun Cho, Sean Welleck

Queering AI Ethics Pedagogy and Practice
Luke Stark, Blake W Hawkins

7:00 AM–8:00 PM | East Hall C
Women in Machine Learning
Co-organizer: Forough Poursabzi-Sangdeh

An all-in-one network for dehazing and beyond
Boyi Li, Xiulian Peng, Zhangyang Wang, Jizheng Xu, Dan Feng

Debunking Debiasing: A Critique of Bias Measurement in NLP
Su Lin Blodgett, Hanna Wallach, Hal Daumé III, Solon Barocas

Efficient Data Deletion from Learned Models and Privacy Implications
Mary Anne Smart, James Zou, Zachary Izzo, Kamalika Chaudhuri

Identification of Patterns in Cystic Fibrosis Physiotheraphy with Unsupervised Learning
Olga Liakhovich, Mihaela Curmei, Tempest van Schaik, Bianca Furtuna, Eleanor Main, Emma Raywood, Nicole Filipow, Kunal Kapoor, Helen Douglas

Industrial Audio Classification with Music Domain Features
Patricia A Ryan, Chenhao Yang

Multi-model Deep Networks for Metastatic Cancert Detection using Biopsy Lymph Node Images
Azadeh Mobasher, Amin Mobasher

Thursday, December 12

7:00 PM–10:00 PM | West 220 – 222
{Dis}Ability in AI
Panelist: Costis Daskalakis, Katherine Heller, Emtiyaz Khan, Hugo Larochelle, Negar Rostamzadeh, Hanna Wallach

Friday, December 13

8:00 AM–6:00 PM | West 215 + 216
CiML 2019: Machine Learning Competitions for All
Adrienne Mendrik, Wei-Wei Tu, Isabelle Guyon, Evelyne Viegas

8:00 AM–6:00 PM | West Exhibition Hall A
Graph Representation Learning

Building Dynamic Knowledge Graphs from Text-based Games
Mikuláš Zelinka, Xingdi Yuan, Marc-Alexandre Côté, Romain Laroche, Adam Trischler

8:20 AM–6:30 PM | West 223 – 224
Human-Centric Machine Learning

Do Machine Teachers Dream of Algorithms?
Gonzalo Ramos, Christopher Meek, Jina Suh, Soroush Ghorashi, Felicia Ng, Nicole Sultanum

Weight of Evidence as a Basis for Human-Oriented Explanations
David Alvarez-Melis, Hal Daumé III, Jennifer Wortman Vaughan, Hanna Wallach

What Is a Proxy and Why Is It a Problem?
Margarita Boyarskaya, Solon Barocas, Hanna Wallach

8:00 AM–6:00 PM | West 109 + 110
KR2ML – Knowledge Representation & Reasoning Meets Machine Learning

Channel Decomposition into Painting Actions
Shih-Chieh Su

Mind The Facts: Knowledge-Boosted Coherent Abstractive Text Summarization
Beliz Gunel, Chenguang Zhu, Michael Zeng, Xuedong Huang

TP-N2F: Tensor Product Representation for Natural To Formal Language Generation
Kezhen Chen, Qiuyuan Huang, Hamid Palangi, Paul Smolensky, Kenneth D. Forbus, Jianfeng Gao

8:00 AM–6:40 PM | West Ballroom A
Machine Learning for Health (ML4H): What makes machine learning in medicine different?
Andrew Beam, Tristan Naumann, Brett Beaulieu-Jones, Madalina Fiterau, Irene Y Chen, Samuel G Finlayson, Emily Alsentzer
Co-organizer: Stephanie Hyland
Senior Advisory Committee Member: Tristan Naumann

8:00 AM–6:00 PM | West 121 + 122
ML for the Developing World (ML4D)

Risks of Using Non-verified Open Data: A case study on using Machine Learning techniques for predicting Pregnancy Outcomes in India
Anusua Trivedi, Sumit Mukherjee, Edmund Tse, Anne Ewing, Juan Lavista Ferres

8:00 AM–6:00 PM | East Meeting Rooms 11 + 12
MLSys: Workshop on Systems for ML
Aparna Lakshmiratan, Siddhartha Sen, Joseph Gonzalez, Dan Crankshaw, Sarah Bird

Compiling Classical ML Pipelines into Tensor Computations for One-size-fits-all Prediction Serving
Supun Nakandala, Gyeong-In Yu, Markus Weimer, Matteo Interlandi

8:00 AM–6:00 PM | West 114 + 115
Retrospectives: A Venue for Self-Reflection in ML Research
Ryan Lowe, Yoshua Bengio, Joelle Pineau, Michela Paganini, Jessica Forde, Shagun Sodhani, Abhishek Gupta, Joel Lehman, Peter Henderson, Kanika Madan

8:00 AM–6:00 PM | West 205 – 207
Robust AI in Financial Services: Data, Fairness, Explainability, Trustworthiness, and Privacy

The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
Solon Barocas, Andrew Selbst, Manish Raghavan

8:00 AM–6:40 AM | East Ballroom A
Safety and Robustness in Decision-making

Airborne Collision Avoidance Systems with Probabilistic Safety Barrier Certificates
Wenhao Luo, Ashish Kapoor

Saturday, December 14

8:00 AM–6:00 PM | East Meeting Rooms 11 + 12
AI for Social Good
Fei Fang, Joseph Bullock, Marc-Antoine Dilhac, Brian P Green, Natalie Saltiel, Dhaval Adjodah, Jack Clark, Sean McGregor, Margaux Luck, Jonathan Penn, Tristan Sylvain, Geneviève Boucher, Sydney Swaine-Simon, Girmaw Abebe Tadesse, Myriam Côté, Anna Bethke, Yoshua Bengio, Abhishek Gupta

8:00 AM–6:30 PM | West Exhibition Hall A
Bridging Game Theory and Deep Learning

Collaborative Machine Learning Markets
Olga Ohrimenko, Shruti Tople, Sebastian Tschiatschek

8:00 AM–6:00 PM | West 217 – 219
Context and Compositionality in Biological and Artificial Neural Systems

Factoring Content and Form in Explicitly Compositional Vector Representations
Paul Smolensky

Enhancing the Transformer with Explicit Relational Encoding for Math Problem Solving
Imanol Schlag, Paul Smolensky, Roland Fernandez, Nebojsa Jojic, Jürgen Schmidhuber, Jianfeng Gao

Uncovering the compositional structure of vector representations with Role Learning Networks
Paul Soulos, R. Thomas McCoy, Tal Linzen, Paul Smolensky

8:00 AM–6:00 PM | West Ballroom C
“Do the right thing”: machine learning and causal inference for improved decision making
Michele Santacatterina, Thorsten Joachims, Nathan Kallus, Adith Swaminathan, David Sontag, Angela Zhou

Causal Transfer Random Forest: Leveraging Observational and Randomization Studies
Shuxi Zeng, Emre Kıcıman, Denis Charles, Joel Pfeiffer, Murat Ali Bayir

Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers
Divyat Mahajan, Amit Sharma

Robust Neural Network for Causal Invariant Features Extraction
Shuxi Zeng, Pengchuan Zhang, Denis Charles, Eren Manavoglu, Emre Kıcıman

Using a causal inference approach to measure the impact of a customer’s attribute on product retention A case study on Microsoft Teams collaborative app
Marie-Laure Charpignon, Liz Manrao, Kamal Choudhary, John Hoegger, MinKyoung Kang

8:00 AM–6:00 PM | West 208 + 209
Document Intelligence
Nigel Duffy, Rama Akkiraju, Tania Bedrax Weiss, Paul Bennett, Hamid Reza Motahari-Nezhad

Learning Structure for Text Generation
Asli Celikyilmaz

8:00 AM–7:00 PM | West Exhibition Hall C
Deep Reinforcement Learning

DRIFT: Deep Reinforcement Learning for Functional Software Testing
Luke Harries, Rebekah Clarke, Timothy Chapman, Swamy Nallamalli, Levent Ozgur, Shuktika Jain, Alex Leung, Steve Lim, Aaron Dietrich, Jose Miguel Hernandez-Lobato, Tom Ellis, Cheng Zhang, Kamil Ciosek

Interactive Fiction Games: A Colossal Adventure
Matthew Hausknecht, Prithviraj V Ammanabrolu, Marc-Alexandre Côté, Xingdi Yuan

9:00 AM–6:00 PM | West 118 – 120
Emergent Communication: Towards Natural Language

Playing log(N)-Questions over Sentences
Peter Potash, Kaheer Suleman

8:00 AM–6:00 PM | East Ballroom B
Fair ML in Healthcare
Senior Advisory Committee Member: Tristan Naumann

8:00 AM–6:00 PM | West 211 – 214
Learning Transferable Skills
Invited Speaker: Katja Hofmann

8:00 AM–6:00 PM | East 1 – 3
Machine Learning for Autonomous Driving

Multi-Agent Connected Autonomous Driving using Deep Reinforcement Learning
Praveen Palanisamy

8:00 AM–6:00 PM | West Ballroom B
Machine Learning with Guarantees

No-Regret and Incentive-Compatible Prediction with Expert Advice
Rupert Freeman, David Pennock, Chara Podimata, Jennifer Wortman Vaughan

8:00 AM–6:45 PM | West 301 – 305
Medical Imaging meets NeurIPS

Binary Mode Multinomial Deep Learning Model for more efficient Automated Diabetic Retinopathy Detection
Anusua Trivedi, J. Desbiens, Ron Gross, S. Gupta, Juan Lavista Ferres, Rahul Dodhia

8:00 AM–6:00 PM | West 202 – 204
ML For Systems

Reinforcement learning for bandwidth estimation and congestion control in real-time communications
Joyce Fang, Martin Ellis, Bin Li, Siyao Liu, Yasaman Hosseinkashi, Michael Revow, Albert Sadovnikov, Ziyuan Liu, Peng Cheng, Sachin Ashok, David Zhao, Ross Cutler, Yan Lu, Johannes Gehrke

8:00 AM–6:00 PM | West 121 – 122
Science meets Engineering of Deep Learning

Non Gaussianity of Stochastic Gradient Noise
Abhishek Panigrahi, Raghav Somani, Navin Goyal, Praneeth Netrapalli

8:00 AM–6:00 PM | East Ballroom C
Tackling Climate Change with ML
David Rolnick, Alexandre Lacoste, Tegan Maharaj, Priya Donti, Lynn Kaack, John Platt, Jennifer Chayes, Yoshua Bengio
Invited Speaker: Lester Mackey

Enhancing Stratospheric Weather Analyses and Forecasts by Deploying Sensors from a Weather Balloon
Kiwan Maeng, Iskender Kushan, Brandon Lucia, Ashish Kapoor

Helping Reduce Environmental Impact of Aviation with Machine Learning
Ashish Kapoor

Panel: Climate Change and AI
Yoshua Bengio, Andrew Ng, Carla Gomes, Lester Mackey, Jeff Dean

8:00 AM–6:00 PM | West 205 – 207
The third Conversational AI workshop – today’s practice and tomorrow’s potential
Alborz Geramifard, Jason Williams, Matt Henderson, Luis Lastras, Dilek Hakkani-Tur, Mari Ostendorf, Milica Gasic, Asli Celikyilmaz

9:00 AM–11:00 AM | West 116 – 117 | Workshop – Competition Track Day 2
The MineRL Competition
Co-organizers: William H. Guss, Mario Ynocente Castro, Cayden Codel, Katja Hofmann, Brandon Houghton, Noboru Kuno, Crissman Loomis, Keisuke Nakata, Stephanie Milani, Sharada Mohanty, Diego Perez Liebana, Ruslan Salakhutdinov, Shinya Shiroshita, Nicholay Topin, Avinash Ummadisingu, Manuela Veloso, Phillip Wang

4:15 PM | West 116 – 117 | Competition
Game of Drones Competition
Matthew Brown, Guada Casuso, Eric Cristofalo, Darius Garza, Nicholas Gyde, Ashish Kapoor, Ratnesh Madaan, Keiko Nagami, Jim Piavis, Davide Scaramuzza, Mac Schwager, Tim Taubner, Sai Vemprala

Posters

Tuesday, December 10

10:45 AM–12:45 PM | East Exhibition Hall B + C #92
Locally Private Gaussian Estimation
Matthew Joseph, Janardhan Kulkarni, Jieming Mao, Steven Wu

10:45 AM–12:45 PM | East Exhibition Hall B + C #150
Can SGD Learn Recurrent Neural Networks with Provable Generalization?
Zeyuan Allen-Zhu, Yuanzhi Li

10:45 AM–12:45 PM | East Exhibition Hall B + C #183
Deep Generalized Method of Moments for Instrumental Variable Analysis
Andrew Bennett, Nathan Kallus, Tobias Schnabel

10:45 AM–12:45 PM | East Exhibition Hall B + C #185
Machine Learning Estimation of Heterogeneous Treatment Effects with Instruments
Vasilis Syrgkanis, Victor Lei, Miruna Oprescu, Maggie Hei, Keith Battocchi, Greg Lewis

10:45 AM–12:45 PM | East Exhibition Hall B + C #198
Successor Uncertainties: Exploration and Uncertainty in Temporal Difference Learning
David Janz, Jiri Hron, Przemysław Mazur, Katja Hofmann, Jose Miguel Hernandez-LobatoSebastian Tschiatschek

5:30 PM–7:30 PM | East Exhibition Hall B + C #87
Twin Auxilary Classifiers GAN
Mingming Gong, Yanwu Xu, Chunyuan Li, Kun Zhang, Kayhan Batmanghelich

5:30 PM–7:30 PM | East Exhibition Hall B + C #92
Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers
Zeyuan Allen-Zhu, Yuanzhi Li, Yingyu Liang

5:30 PM–7:30 PM | East Exhibition Hall B + C #135
Integrating mechanistic and structural causal models enables counterfactual inference in complex systems
Robert Ness, Kaushal Paneri, Olga Vitek

5:30 PM–7:30 PM | East Exhibition Hall B + C #154
Poisson-randomized Gamma Dynamical Systems
Aaron Schein, Scott Linderman, Mingyuan Zhou, David Blei, Hanna Wallach

5:30 PM–7:30 PM | East Exhibition Hall B + C #179
Better Exploration with Optimistic Actor Critic
Kamil Ciosek, Quan Vuong, Robert Loftin, Katja Hofmann

5:30 PM–7:30 PM | East Exhibition Hall B + C #213
Staying up to Date with Online Content Changes Using Reinforcement Learning for Scheduling
Andrey Kolobov, Yuval Peres, Cheng Lu, Eric Horvitz

Wednesday, December 11

10:45 AM–12:45 PM | East Exhibition Hall B + C #104
Normalization Helps Training of Quantized LSTM
Lu Hou, Jinhua Zhu, James Kwok, Fei Gao, Tao Qin, Tie-Yan Liu

10:45 AM–12:45 PM | East Exhibition Hall B + C #107
Shallow RNN: Accurate Time-series Classification on Resource Constrained Devices
Don Dennis, Durmus Alp Emre Acar, Vikram Mandikal, Vinu Sankar Sadasivan, Venkatesh Saligrama, Harsha Vardhan Simhadri, Prateek Jain

10:45 AM–12:45 PM | East Exhibition Hall B + C #158
A Stochastic Composite Gradient Method with Incremental Variance Reduction
Junyu Zhang, Lin Xiao

10:45 AM–12:45 PM | East Exhibition Hall B + C #163
Understanding the Role of Momentum in Stochastic Gradient Methods
Igor Gitman, Hunter Lang, Pengchuan Zhang, Lin Xiao

10:45 AM–12:45 PM | East Exhibition Hall B + C #203
Distributional Reward Decomposition for Reinforcement Learning
Zichuan Lin, Li Zhao, Derek Yang, Tao Qin, Tie-Yan Liu, Guangwen Yang

10:45 AM–12:45 PM | East Exhibition Hall B + C #208
Fully Parameterized Quantile Function for Distributional Reinforcement Learning
Derek Yang, Li Zhao, Zichuan Lin, Tao Qin, Jiang Bian, Tie-Yan Liu

10:45 AM–12:45 PM | East Exhibition Hall B + C #231
What Can ResNet Learn Efficiently, Going Beyond Kernels?
Zeyuan Allen-Zhu, Yuanzhi Li

5:00 PM–7:00 PM | East Exhibition Hall B + C #3
Low-Rank Bandit Methods for High-Dimensional Dynamic Pricing
Jonas Mueller, Vasilis Syrgkanis, Matt Taddy

5:00 PM–7:00 PM | East Exhibition Hall B + C #30
Metalearned Neural Memory
Tsendsuren Munkhdalai, Alessandro Sordoni, Tong Wang, Adam Trischler

5:00 PM–7:00 PM | East Exhibition Hall B + C #76
FastSpeech: Fast, Robust and Controllable Text to Speech
Yi Ren, Yangjun Ruan, Xu Tan, Tao Qin, Sheng Zhao, Zhou Zhao, Tie-Yan Liu

5:00 PM–7:00 PM | East Exhibition Hall B + C #94
Rand-NSG: Fast Accurate Billion-point Nearest Neighbor Search on a Single Node
Suhas Jayaram Subramanya, Devvrit Lnu, Harsha Vardhan Simhadri, Ravishankar Krishnawamy, Rohan Kadekodi

5:00 PM–7:00 PM | East Exhibition Hall B + C #101
A Tensorized Transformer for Language Modeling Xindian Ma, Peng Zhang, Shuai Zhang, Nan Duan, Yuexian Hou, Ming Zhou, Dawei Song

5:00 PM–7:00 PM | East Exhibition Hall B + C #106
Improving Textual Network Learning with Variational Homophilic Embeddings
Wenlin Wang, Chenyang Tao, Zhe Gan, Guoyin Wang, Liqun Chen, Xinyuan Zhang, Ruiyi Zhang, Qian Yang, Ricardo Henao, Lawrence Carin

5:00 PM–7:00 PM | East Exhibition Hall B + C #122
Neuropathic Pain Diagnosis Simulator for Causal Discovery Algorithm Evaluation
Ruibo Tu, Kun Zhang, Bo Bertilson, Hedvig Kjellstrom, Cheng Zhang

5:00 PM–7:00 PM | East Exhibition Hall B + C #136
Bias Correction of Learned Generative Models using Likelihood-Free Importance Weighting
Aditya Grover, Jiaming Song, Ashish Kapoor, Kenneth Tran, Alekh Agarwal, Eric Horvitz, Stefano Ermon

5:00 PM–7:00 PM | East Exhibition Hall B + C #143
Icebreaker: Element-wise Efficient Information Acquisition with Active Learning
Wenbo Gong, Sebastian Tschiatschek, Sebastian Nowozin, Richard E Turner, José Miguel Hernández-Lobato, Cheng Zhang

5:00 PM–7:00 PM | East Exhibition Hall B + C #202
Budgeted Reinforcement Learing in Continuous State Space
Nicolas Carrara, Edouard Leurent, Romain Laroche, Tanguy Urvoy, Odalric-Ambrym Maillard, and Olivier Pietquin

5:00 PM–7:00 PM | East Exhibition Hall B + C #232
Semi-Parametric Efficient Policy Learning with Continuous Actions
Victor Chernozhukov, Mert Demirer, Greg Lewis, Vasilis Syrgkanis

5:00 PM–7:00 PM | East Exhibition Hall B + C #242
Wide Feedforward or Recurrent Neural Networks of Any Architecture are Gaussian Processes
Greg Yang

Thursday, December 12

10:45 AM–12:45 PM | East Exhibition Hall B + C #21
Policy Poisoning in Batch Reinforcement Learning and Control
Yuzhe Ma, Xuezhou Zhang, Wen Sun, Jerry Zhu

10:45 AM–12:45 PM | East Exhibition Hall B + C #24
Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers
Hadi Salman, Jerry Li, Ilya Razenshteyn, Pengchuan Zhang, Huan Zhang, Sebastien Bubeck, Greg Yang

10:45 AM–12:45 PM | East Exhibition Hall B + C #27
Efficient Forward Architecture Search
Hanzhang Hu, John Langford, Rich Caruana, Saurajit Mukherjee, Eric Horvitz, Debadeepta Dey

10:45 AM–12:45 PM | East Exhibition Hall B + C #34
Procrastinating with Confidence: Near-Optimal, Anytime, Adaptive Algorithm Configuration
Robert Kleinberg, Kevin Leyton-Brown, Brendan Lucier, Devon Graham

10:45 AM–12:45 PM | East Exhibition Hall B + C #70
Unsupervised State Representation Learning in Atari
Ankesh Anand, Sherjil Ozair, Evan Racah, Yoshua Bengio, Marc-Alexandre CôtéDevon Hjelm

10:45 AM–12:45 PM | East Exhibition Hall B + C #75
Characterizing Bias in Classifiers using Generative Models
Daniel McDuff, Shuang Ma, Yale Song, Ashish Kapoor

10:45 AM–12:45 PM | East Exhibition Hall B + C #169
On the Convergence Rate of Training Recurrent Neural Networks
Zeyuan Allen-Zhu, Yuanzhi Li, Zhao Song

10:45 AM–12:45 PM | East Exhibition Hall B + C #173
Using Statistics to Automate Stochastic Optimization
Hunter Lang, Pengchuan Zhang, Lin Xiao

10:45 AM–12:45 PM | East Exhibition Hall B + C #210
The Step Decay Schedule: A Near Optimal, Geometrically Decaying Learning Rate Procedure For Least Squares
Rong Ge, Sham M. Kakade, Rahul Kidambi, Praneeth Netrapalli

10:45 AM–12:45 PM | East Exhibition Hall B + C #228
Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck
Maximilian Igl, Kamil CiosekYingzhen LiSebastian TschiatschekCheng ZhangSam DevlinKatja Hofmann

10:45 AM–12:45 PM | East Exhibition Hall B + C #232
Reinforcement Learning with Convex Constraints
Sobhan Miryoosefi, Kianté Brantley, Hal Daume III, Miro Dudik, Robert Schapire

5:00 PM–7:00 PM | East Exhibition Hall B + C #9
Minimum Stein Discrepancy Estimators
Alessandro Barp, Francois-Xavier Briol, Andrew Duncan, Mark Girolami, Lester Mackey

5:00 PM–7:00 PM | East Exhibition Hall B + C #32
Learning Representations by Maximizing Mutual Information Across Views
Philip Bachman, R Devon Hjelm, William Buchwalter

5:00 PM–7:00 PM | East Exhibition Hall B + C #51
Sample Complexity of Learning Mixture of Sparse Linear Regressions
Akshay Krishnamurthy, Arya Mazumdar, Andrew McGregor, Soumyabrata Pal

5:00 PM–7:00 PM | East Exhibition Hall B + C #140
Unified Language Model Pre-training for Natural Language Understanding and Generation
Li Dong, Nan Yang, Wenhui Wang, Furu Wei, Xiaodong Liu, Yu Wang, Jianfeng Gao, Ming Zhou, Hsiao-Wuen Hon

5:00 PM–7:00 PM | East Exhibition Hall B + C #142
Adaptive Influence Maximization with Myopic Feedback
Binghui Peng, Wei Chen

5:00 PM–7:00 PM | East Exhibition Hall B + C #152
A Convex Relaxation Barrier to Tight Robustness Verification of Neural Networks
Hadi Salman, Greg Yang, Huan Zhang, Cho-Jui Hsieh, Pengchuan Zhang

5:00 PM–7:00 PM | East Exhibition Hall B + C #153
An Algorithmic Framework For Differentially Private Data Analysis on Trusted Processors
Joshua Allen, Bolin Ding, Janardhan Kulkarni, Harsha Nori, Olga Ohrimenko, Sergey Yekhanin

5:00 PM–7:00 PM | East Exhibition Hall B + C #160
Oblivious Sampling Algorithms for Private Data Analysis
Sajin Sasy, Olga Ohrimenko

5:00 PM–7:00 PM | East Exhibition Hall B + C #168
Program Synthesis and Semantic Parsing with Learned Code Idioms
Richard Shin, Miltos Allamanis, Marc Brockschmidt, Alex Polozov

5:00 PM–7:00 PM | East Exhibition Hall B + C #194
Ordered Memory
Yikang Shen, Shawn Tan, Arian Hosseini, Zhouhan Lin, Alessandro Sordoni, Aaron Courville

5:00 PM–7:00 PM | East Exhibition Hall B + C #213
Efficient Algorithms for Smooth Minimax Optimization
Kiran Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh

Spotlights + Oral presentations

Oral presentations

Tuesday, December 10

4:50 PM–5:05 PM | West Ballrooms A + B
Using a Logarithmic Mapping to Enable Lower Discount Factors in Reinforcement Learning
Harm van SeijenMehdi Fatemi, Arash Tavakoli

Spotlight sessions

Tuesday, December 10

10:40 AM–10:45 AM | West Exhibition Hall A
Quantum Entropy Scoring for Fast Robust Mean Estimation and Improved Outlier Detection
Yihe Dong, Samuel B. Hopkins, Jerry Li

5:05 PM–5:10 PM | West Ballroom A + B
Better Exploration with Optimistic Actor Critic
Kamil Ciosek, Quan Vuong, Robert Loftin, Katja Hofmann

Wednesday, December 11

4:20 PM–4:25 PM | West Exhibition Hall A
Model selection for contextual bandits
Dylan Foster, Akshay Krishnamurthy, Haipeng Luo

Thursday, December 12

10:20 AM–10:25 AM | West Exhibition Hall A
Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers
Hadi Salman, Greg Yang, Jerry Li, Pengchuan Zhang, Huan Zhang, Ilya Razenshteyn, Sebastien Bubeck

Demos

Come by our booth (#301) to see demos of our latest research and meet with our team! See schedule below:

Monday, December 9

Time Demos Meet Our Team
10:30 AM–11:15 AM
  • Responsible AI
  • Infer.NET
  • Andrew Fitzgibbon – Vision
  • Anna Mitenkova – RL
1:15 PM–2:00 PM
  • BERT on Azure ML
  • Distributional Reward Decomposition for Reinforcement Learning
  • Ali Zaidi – NLP, Stats
  • Kamil Ciosek – RL
  • Shawn Jain – AI Residency
2:00 PM–2:45 PM
  • Metalearned Neural Memory
  • Fully Parameterized Quantile Function for Distributional Reinforcement Learning
  • Harm van Seijen – RL
  • Ravi Pandya – Health
4:45 PM–5:00 PM
  • BERT on Azure ML
  • Icebreaker: Efficient Information Acquisition with Active Learning
6:30 PM–7:30 PM
  • Alex Polozov – Program synthesis, NL2Code
  • Jan Stühmer – Generative Models, Vision
  • Keiji Kanazawa – Azure Machine Learning
  • Marc Brockschmidt – GNNs
6:35 PM–8:30 PM
  • IPU-Accelerated Medical Imaging on Microsoft Azure

Tuesday, December 10

Time Demos Meet Our Team
9:20 AM–10:05 AM
  • Responsible AI: interpretability & fairness in machine learning
  • BERT on Azure ML
  • Abhishek Rao – Text classification
  • Ehi Nosakhare – ML or healthcare/interpretability
  • Govert Verkes – Vision
  • John-Mark Agosta – Azure
  • Yi Li – Knowledge graph
12:45 PM–1:30 PM
  • Responsible AI: interpretability & fairness in machine learning
  • Efficient Forward Neural Architecture Search
  • Andrew Fitzgibbon – Vision
  • Danielle Belgrave – AI for Healthcare
  • Kaushal Paneri – Causal Inference
  • Michael Revow – Real time communication, Video
1:30 PM–2:15 PM
  • Responsible AI: interpretability & fairness in machine learning
  • BERT on Azure ML
  • Anna Mitenkova – RL
  • Cassandra Oduola – Deep Learning, CV, HCI
  • Isabel Chien – AI for healthcare
  • Raluca Georgescu – RL
3:25 PM–4:10 PM
  • Responsible AI: interpretability & fairness in machine learning
  • Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers
  • Akshay Krishnamurthy – RL Theory
  • Samira Shabanian – Generative models, FATE
  • Marc Brockschmidt – GNNs
  • Kaushal Paneri – Causal Inference

Wednesday, December 11

Time Demos Meet Our Team
9:20 AM–10:05 AM
  • High performance inferencing for interoperable ONNX models
  • Microsoft Rocket Video Analytics Platform
  • Andrew Fitzgibbon – Vision
  • Anuj Bhatia – Azure
  • Javier Alvarez – AI for healthcare
  • Yuanchao Shu – Video and Systems
12:45 PM–1:30 PM
  • High performance inferencing for interoperable ONNX models
  • Using a Logarithmic Mapping to Enable Lower Discount Factors in Reinforcement Learning
  • Forough Poursabzi-Sangdeh – FATE, CSS
  • Lindsey Allen – AI Platform
  • Siyu Yang – AI for Good
  • Yi Li – Knowledge graph
  • Yixi Xu – AI for Good
1:30 PM–2:15 PM
  • High performance inferencing for interoperable ONNX models
  • FastSpeech: Fast, Robust and Controllable Text to Speech
  • Danielle Belgrave – AI for Healthcare
  • Judy Shen – AI Residency
  • Lindsey Allen – AI Platform
  • Marc Brockschmidt – GNNs
  • Raluca Georgescu – RL
3:05 PM–3:50 PM
  • Responsible AI: interpretability & fairness in machine learning
  • Infer.NET
  • Alex Polozov – Program synthesis, NL2Code
  • Allison Hegel – AI Residency
  • Ehi Nosakhare – MAIDAP/ ML for health/ Interpretability
  • Forough Poursabzi-Sangdeh – FATE, CSS
  • Lindsey Allen – AI Platform

#AlchemyFriends

Print your own copy of Alchemy with Friends to play at home. Share your favorite card combinations using #AlchemyFriends on Twitter, Facebook, or Instagram.

Animated illustration of how to play #AlchemyFriends

Microsoft Research blog

Researcher Tools

Image showing rectangles of various sizes passing through a magic door and becoming same size to depict logarithmic mapping.

Logarithmic mapping allows for low discount factors by creating action gaps similar in size

While reinforcement learning (RL) has seen significant successes over the past few years, modern deep RL methods are often criticized for how sensitive they are with respect to their hyper-parameters. One such hyper-parameter is the discount factor, which controls how future rewards are weighted compared to immediate rewards. The objective that one wants to optimize in RL is often best described as an undiscounted sum of rewards (for example, maximizing the total score in a…

November 2019

Microsoft Research Blog

Optimistic Actor Critic avoids the pitfalls of greedy exploration in reinforcement learning

One of the core directions of Project Malmo is to develop AI capable of rich interactions. Whether that means learning new skills to apply to challenging problems, understanding complex environments, or knowing when to enlist the help of humans, reinforcement learning (RL) is a core enabling technology for building these types of AI. In order to perform RL well, agents need to do exploration efficiently, which means understanding when to try new things out and…

November 2019

Microsoft Research Blog

Graphic showing the components of the Icebreaker model

Icebreaker: New model with novel element-wise information acquisition method reduces cost and data needed to train machine learning models

In many real-life scenarios, obtaining information is costly, and getting fully observed data is almost impossible. For example, in the recruiting world, obtaining relevant information (in other words, a feature value) for a company could mean performing time-consuming interviews. The same applies to many other scenarios, such as in education and the medical field, where each feature value is an often more complex answer to a question. Unfortunately, AI-aided decision making usually requires large amounts…

November 2019

Microsoft Research Blog

Image of an AI agent finding a cup of coffee

The road less traveled: With Successor Uncertainties, RL agents become better informed explorers

Imagine moving to a new city. You want to get from your new home to your new job. Unfamiliar with the area, you ask your co-workers for the best route, and as far as you can tell … they’re right! You get to work and back easily. But as you acclimate, you begin to wonder: Is there a more scenic route, perhaps, or a route that passes by a good coffee spot? The fundamental question…

December 2019

Microsoft Research Blog

Gurdeep Pall and Ashish Kapoor on the Microsoft Research Podcast

Autonomous systems, aerial robotics and Game of Drones with Gurdeep Pall and Dr. Ashish Kapoor

Episode 100 | November 27, 2019 – There’s a lot of excitement around self-driving cars, delivery drones, and other intelligent, autonomous systems, but before they can be deployed at scale, they need to be both reliable and safe. That’s why Gurdeep Pall, CVP of Business AI at Microsoft, and Dr. Ashish Kapoor, who leads research in Aerial Informatics and Robotics, are using a simulated environment called AirSim to reduce the time, cost and risk of…

November 2019

Microsoft Research Blog

Metalearned Neural Memory: Teaching neural networks how to remember

Memory is an important part of human intelligence and the human experience. It grounds us in the current moment, helping us understand where we are and, consequently, what we should do next. Consider the simple example of reading a book. The ultimate goal is to understand the story, and memory is the reason we’re able to do so. Memory allows us to efficiently store the information we encounter and later recall the details we’ve previously…

December 2019

Microsoft Research Blog

illustrated palm tree on an island

Provable guarantees come to the rescue to break attack-defense cycle in adversarial machine learning

Artificial intelligence has evolved to become a revolutionary technology. It is rapidly changing the economy, both by creating new opportunities (it’s the backbone of the gig economy) and by bringing venerable institutions, like transportation, into the 21st century. Yet deep at its core something is amiss, and more and more experts are worried: the technology seems to be extremely brittle, a phenomenon epitomized by adversarial examples. Adversarial examples exploit weaknesses in modern AI. Today, most…

December 2019

Microsoft Research Blog

Game of Drones simulation

Game of Drones at NeurIPS 2019: Simulation-based drone-racing competition built on AirSim

Drone racing has transformed from a niche activity sparked by enthusiastic hobbyists to an internationally televised sport. In parallel, computer vision and machine learning are making rapid progress, along with advances in agile trajectory planning, control, and state estimation for quadcopters. These advances enable increased autonomy and reliability for drones. More recently, the unmanned aerial vehicle (UAV) research community has begun to tackle the drone-racing problem. This has given rise to competitions, with the goal…

December 2019

Microsoft Research Blog

Project Petridish: Efficient forward neural architecture search

Having experience in deep learning doesn’t hurt when it comes to the often mysterious, time- and cost-consuming process of hunting down an appropriate neural architecture. But truth be told, no one really knows what works the best on a new dataset and task. Relying on well-known, top-performing networks provides few guarantees in a space where your dataset can look very different from anything those proven networks have encountered before. For example, a network that worked…

December 2019

Microsoft Research Blog

FastSpeech: New text-to-speech model improves on speed, accuracy, and controllability

FastSpeech: New text-to-speech model improves on speed, accuracy, and controllability

Text to speech (TTS) has attracted a lot of attention recently due to advancements in deep learning. Neural network-based TTS models (such as Tacotron 2, DeepVoice 3 and Transformer TTS) have outperformed conventional concatenative and statistical parametric approaches in terms of speech quality. Neural network-based TTS models usually first generate a mel-scale spectrogram (or mel-spectrogram) autoregressively from text input and then synthesize speech from the mel-spectrogram using a vocoder. (Note: the Mel scale is used…

December 2019

Microsoft Research Blog

an equation where x and y are unknowns above an illustration with x and y bouncing through like pinballs

Next-generation architectures bridge gap between neural and symbolic representations with neural symbols

In both language and mathematics, symbols and their mutual relationships play a central role. The equation x = 1/y asserts the symbols x and y—that is, what they stand for—are related reciprocally; Kim saw the movie asserts that Kim and the movie are perceiver and stimulus. People are extremely adept with the symbols of language and, with training, become adept with the symbols of mathematics. For many decades, cognitive science explained these human abilities by…

December 2019

Microsoft Research Blog

animation of reinforcement learning agents beating human competitors in Atari

Finding the best learning targets automatically: Fully Parameterized Quantile Function for distributional RL

Reinforcement learning has achieved great success in game scenarios, with RL agents beating human competitors in such games as Go and poker. Distributional reinforcement learning, in particular, has proven to be an effective approach for training an agent to maximize reward, producing state-of-the-art results on Atari games, which are widely used as benchmarks for testing RL algorithms. Because of the intrinsic randomness of game environments—with the roll of the dice in Monopoly, for example, you…

December 2019

Microsoft Research Blog

Downloads

Videos