Mila Researchers prep for NeurIPS 2020
Earlier this October, the Neural Information Processing Systems Conference (NeurIPS) unveiled the complete list of accepted publications for its 34th edition, which will be entirely virtual.
As one of the largest and most important machine learning conferences in the world, the research culture around the submission process has become increasingly competitive. To give an example, the NeurIPS 2020 Chairs have announced that of the 12,115 abstracts submitted, a total of 1,903 have been accepted for this year’s conference.
In light of the increasing pressure our researchers are facing, it is important to commemorate all the work our students and professors have done in producing their research and organizing workshops for this conference. Accepted papers from our community cover a range of topics including, but not limited to, novel continual model-agnostic meta-learning approaches, solutions to oversmoothness in graph convolutional networks, and two policy regularization methods in multi-agent deep reinforcement learning.
Below is the complete list of publications, as well as an overview of the workshops co-organized by Mila members:
- Untangling tradeoffs between recurrence and self-attention in neural networks
Giancarlo Kerg, Bhargav Kanuparthi, Anirudh Goyal, Kyle Goyette, Yoshua Bengio, Guillaume Lajoie
https://arxiv.org/abs/2006.09471 - Top-k Training of GANs: Improving GAN Performance by Throwing Away Bad Samples
Samarth Sinha, Zhengli Zhao, Anirudh Goyal, Colin Raffel, Augustus Odena
https://arxiv.org/abs/2002.06224 - Hybrid Models for Learning to Branch
Prateek Gupta, Maxime Gasse, Elias B. Khalil, M. Pawan Kumar, Andrea Lodi, Yoshua Bengio
https://arxiv.org/abs/2006.15212 - The LoCA Regret: A Consistent Metric to Evaluate Model-Based Behavior in Reinforcement Learning
Harm van Seijen, Hadi Nekoei, Evan Racah, Sarath Chandar
https://arxiv.org/abs/2007.03158 - Your GAN is Secretly an Energy-based Model and You Should use Discriminator Latent Sampling
Tong Che, Ruixiang Zhang, Jascha Sohl-Dickstein, Hugo Larochelle, Liam Paull, Yuan Cao, Yoshua Bengio
https://arxiv.org/abs/2003.06060 - Counterexample-Guided Learning of Monotonic Neural Networks
Aishwarya Sivaraman, Golnoosh Farnadi, Todd Millstein, Guy Van den Broeck
https://arxiv.org/abs/2006.08852 - Adversarial Example Games
Joey Bose, Gauthier Gidel, Hugo Berard, Andre Cianflone, Pascal Vincent, Simon Lacoste-Julien, William Hamilton
https://arxiv.org/abs/2007.00720 - Online Fast Adaptation and Knowledge Accumulation: A New Approach to Continual Learning
Massimo Caccia, Pau Rodriguez, Oleksiy Ostapenko, Fabrice Normandin, Min Lin, Lucas Caccia, Issam Laradji, Irina Rish, Alexandre Lacoste, David Vazquez, Laurent Charlin
https://arxiv.org/abs/2003.05856 - Learning Dynamic Belief Graphs to Generalize on Text-Based Games
Ashutosh Adhikari, Xingdi Yuan, Marc-Alexandre Côté, Mikulás Zelinka, Marc-Antoine Rondeau, Romain Laroche, Pascal Poupart, Jian Tang, Adam Trischler, William Hamilton
https://arxiv.org/abs/2002.09127 - Measuring Systematic Generalization in Neural Proof Generation with Transformers
Nicolas Gontier, Koustuv Sinha, Siva Reddy, Christopher Pal
https://arxiv.org/abs/2009.14786 - Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework
Dinghuai Zhang, Mao Ye, Chengyue Gong, Zhanxing Zhu, Qiang Liu
https://arxiv.org/abs/2002.09169 - Forethought and Hindsight in Credit Assignment
Veronica Chelu, Doina Precup, Hado van Hasselt - Real World Games Look like Spinning Tops
Wojciech Marian Czarnecki, Gauthier Gidel, Brendan Tracy, Karl Tuyls, Shayegan Omidshafiefi, David Balduzzi, Max Jaderberg
https://arxiv.org/abs/2004.09468 - Neumann Networks: Differential programming for supervised learning with missing values
Marine Le Morvan, Julie Josse, Thomas Moreau, Erwan Scornet, Gaël Varoquaux
https://arxiv.org/abs/2007.01627 - Training Linear Finite-State Machines
Arash Ardakani, Amir Ardakani, Warren Gross - RL Unplugged: Benchmarks for Offline Reinforcement Learning
Caglar Gulcehre, Ziyu Wang, Alexander Novikov, Tom Le Paine, Sergio Gomez Colmenarejo, Konrad Zolna, Rishabh Agarwal, Josh Merel, Daniel Mankowitz, Cosmin Paduraru, Gabriel Dulac-Arnold, Jerry Li, Mohammad Norouzi, Matt Hoffman, Ofir Nachum, George Tucker, Nicolas Heess, Nando de Freitas
https://arxiv.org/abs/2006.13888 - Graph Policy Network for Transferable Active Learning on Graphs
Shengding Hu, Zheng Xiong, Xingdi Yuan, Marc-Alexandre Côté, Zhiyuan Liu, Jian Tang
https://arxiv.org/abs/2006.13463 - Towards Interpretable Natural Language Understanding with Explanations as Latent Variables
Wangchunshu Zhou, Jinyi Hu, Hanlin Zhang, Xiaodan Liang, Maosong Sun, Chenyan Xiong, Jian Tang - Promoting Coordination through Policy Regularization in Multi-Agent Deep Reinforcement Learning
Julien Roy, Paul Barde, Félix Harvey, Derek Nowrouzezahrai, Christopher Pal
https://arxiv.org/abs/1908.02269 - Scattering GCN: Overcoming Oversmoothness in Graph Convolutional Networks
Yimeng Min, Frederik Wenkel, Guy Wolf
https://arxiv.org/abs/2003.08414 - In search of robust measures of generalization
Gintare Karolina Dziugaite, Alexandre Drouin, Brady Neal, Nitarshan Rajkumar, Ethan Caballero, Linbo Wang, Ioannis Mitliagkas, Daniel Roy - An Equivalence between Loss Functions and Non-Uniform Sampling in Experience Replay
Scott Fujimoto, David Meger, Doina Precup - Learning Graph Structure with A Finite-State Automaton Layer
Daniel D. Johnson, Hugo Larochelle, Daniel Tarlow
https://arxiv.org/abs/2007.04929 - Unsupervised Learning of Dense Visual Representations
Pedro O. Pinheiro, Amjad Almahairi, Ryan Benmalek, Florian Golemo, Aaron Courville - Equivariant Networks for Hierarchical Structures
Renhao Wang, Marjan Albooyeh, Siamak Ravanbakhsh - Symbols: Probing Learning Algorithms with Synthetic Datasets
Alexandre Lacoste, Pau Rodríguez, Frederic Branchaud-Charron, Massimo Caccia, Issam Hadj Laradji, Alexandre Drouin, Matthew Craddock, Laurent Charlin, David Vázquez
https://arxiv.org/abs/2009.06415 - Explicit Regularization is Stronger than Implicit Bias: A Study of SGD around Bad Global Minima
Shengchao Liu, Dimitris Papailiopoulos, Dimitris Achlioptas - Novelty Search in representational space for sample efficient exploration
Ruo Yu Tao, Vincent François-Lavet, Joelle Pineau
https://arxiv.org/abs/2009.13579 - La-MAML: Look-ahead Meta Learning for Continual Learning
Gunshi Gupta, Karmesh Yadav, Liam Paull
https://arxiv.org/abs/2007.13904 - Adversarial Soft Advantage Fitting: Imitation Learning Without Policy Optimization
Paul Barde, Julien Roy, Wonseok Jeon, Joelle Pineau, Christopher Pal, Derek Nowrouzezahrai
https://arxiv.org/abs/2006.13258 - Differentiable Causal Discovery from Interventional Data
Philippe Brouillard, Sébastien Lachapelle, Alexandre Lacoste, Simon Lacoste-Julien, Alexandre Drouin
https://arxiv.org/abs/2007.01754 - Reward Propagation using Graph Convolutional Networks
Martin Klissarov and Doina Precup
https://arxiv.org/abs/2010.02474 - Uncovering the Topology of Time-Varying fMRI Data using Cubical Persistence
Bastian Rieck, Tristan Yates, Christian Bock, Karsten Borgwardt, Guy Wolf, Nicholas Turk-Browne, Smita Krishnaswamy
https://arxiv.org/abs/2006.07882
- Untangling tradeoffs between recurrence and self-attention in neural networks
Workshops:
- Tackling Climate Change with Machine Learning
Tegan Maharaj, Priya Donti, Lynn Kaack, Alexandre Lacoste, Andrew Ng, John Platt, Jennifer Chayes, Yoshua Bengio
https://www.climatechange.ai/events/neurips2019.html - Topological Data Analysis and Beyond
Bastian Rieck, Frederic Chazal, Smita Krishnaswamy, Roland Kwitt, Karthikeyan Natesan Ramamurthy, Yuhei Umeda, and Guy Wolf
https://tda-in-ml.github.io/ - Differential Geometry meets Deep Learning (DiffGeo4DL)
Joey Bose, William Hamilton
https://sites.google.com/view/diffgeo4dl/ - Differentiable vision, graphics, and physics applied to machine learning
Krishna Murthy, Kelsey Allen, Victoria Dean, Johanna Hansen, Shuran Song, Florian Shkurti, Liam Paull, Derek Nowrouzezahrai, Josh Tenenbaum
https://montrealrobotics.ca/diffcvgp/ - Resistance AI
Mattie Tesfaldet, J Khadijah Abdurahman, William Agnew, Abeba Birhane, Elliot Creager, Agata Foryciarz, Pratyusha Ria Kalluri, Sayash Kapoor, Raphael Gontijo Lopes, Manuel Sabin, Marie-Therese Png, Maria Skoularidou, Ramon Vilarino, Rose E. Wang
https://sites.google.com/view/resistance-ai-neurips-20/home - Biological and Artificial Reinforcement Learning
Raymond Chua, Feryal Behbahani, Sara Zannone, Doina Precup, Ida Momennejad, Blake Richards, Rui Ponte Costa
https://sites.google.com/view/biologicalandartificialrl - ML Retrospectives, Surveys & Meta-Analyses
Chhavi Yadav, Prabhu Pradhan, Abhishek Gupta, Peter Henderson, Ryan Lowe, Jessica Forde, Jesse Dodge, Mayoore Jaiswal, Joelle Pineau
https://ml-retrospectives.github.io/neurips2020/ - Object Representations for Learning and Reasoning
William Agnew, Rim Assouel, Michael Chang, Antonia Creswell, Eliza Kosoy, Aravind Rajeswaran, Sjoerd van Steenkiste
https://orlrworkshop.github.io/index.html - Algorithmic Fairness through the Lens of Causality and Interpretability
Golnoosh Farnadi, Awa Dieng, Jessica Schrouff, Matt Kusner, Fernando Diaz
https://www.afciworkshop.org/ - AI for Earth Sciences
Natasha Dudek, S. Karthik Mukkavilli, Johanna Hansen, Kelly Kochanski, S. Karthik Mukkavilli, Tom Beucler, Karthik Kashinath, Mayur Mudigonda, Amy McGovern
https://ai4earthscience.github.io/neurips-2020-workshop/organizers.html - ML4H: Machine Learning for Health
Benjamin Akera, Emily Alsentzer, Oliver J. Bear, Ioana Bica, Irene Chen, Fabian Falck, Stephanie Hyland, Dani Kyasseh, Matthew McDermott, Ehi Nosakhare, Charles Onu, Stephen Pfohl, Aahlad Manas Puli, Suproteem Sarkar, Allen Schmaltz
https://ml4health.github.io/2020/pages/organizers.html - Women in Machine Learning
Amy Zhang, Elizabeth Wood, Mel Woghiren, Kristy Choi, Raquel Aoki, Judy Hanwen Shen, Belén Saldías, Xenia Miscouridou, Krystal Maughan, Xinyi Chen, Erin Grant
https://wimlworkshop.org/neurips2020/contact-us/ - Offline Reinforcement Learning
Rishabh Agarwal, Aviral Kumar, George Tucker, Doina Precup, Lihong Li
https://offline-rl-neurips.github.io/
- Tackling Climate Change with Machine Learning