Stanford AI Lab Papers and Talks at NeurIPS 2021

The thirty-fifth Conference on Neural Information Processing Systems (NeurIPS) 2021 is being hosted virtually from Dec 6th - 14th. We’re excited to share all the work from SAIL that’s being presented at the main conference , at the Datasets and Benchmarks track and the various workshops , and you’ll find links to papers, videos and blogs below. Some of the members in our SAIL community also serve as co-organizers of several exciting workshops that will take place on Dec 13-14, so we hope you will check them out! Feel free to reach out to the contact authors and the workshop organizers directly to learn more about the work that’s happening at Stanford! Main Conference Improving Compositionality of Neural Networks by Decoding Representations to Inputs Authors : Mike Wu, Noah Goodman, Stefano Ermon Contact : wumike@stanford.edu Links: Paper Keywords : generative models, compositionality, decoder Reverse engineering recurrent neural networks with Jacobian switching linear dynamical systems Authors : Jimmy T.H. Smith, Scott W. Linderman, David Sussillo Contact : jsmith14@stanford.edu Links: Paper | Website Keywords : recurrent neural networks, switching linear dynamical systems, interpretability, fixed points Compositional Transformers for Scene Generation Authors : Drew A. Hudson, C. Lawrence Zitnick Contact : dorarad@cs.stanford.edu Links: Paper | Github Keywords : GANs, transformers, compositionality, scene synthesis Combining Recurrent, Convolutional, and Continuous-time Models with Linear State Space Layers Authors : Albert Gu, Isys Johnson, Karan Goel, Khaled Saab, Tri Dao, Atri Rudra, Chris Ré Contact : albertgu@stanford.edu Links: Paper Keywords : recurrent neural networks, rnn, continuous models, state space, long range dependencies, sequence modeling Emergent Communication of Generalizations Authors : Jesse Mu, Noah Goodman Contact : muj@stanford.edu Links: Paper | Video Keywords : emergent communication, multi-agent communication, language grounding, compositionality Deep Learning on a Data Diet: Finding Important Examples Early in Training Authors : Mansheej Paul, Surya Ganguli, Gintare Karolina Dziugaite Contact : mansheej@stanford.edu Links: Paper Keywords : data pruning ELLA: Exploration through Learned Language Abstraction Authors : Suvir Mirchandani, Siddharth Karamcheti, Dorsa Sadigh Contact : suvir@cs.stanford.edu Links: Paper | Video Keywords : instruction following, reward shaping, reinforcement learning CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation Authors : Yusuke Tashiro, Jiaming Song, Yang Song, Stefano Ermon Contact : ytashiro@stanford.edu Links: Paper | Website Keywords : score-based generative modeling, time series imputation Confidence-Aware Imitation Learning from Demonstrations with Varying Optimality Authors : Songyuan Zhang, Zhangjie Cao, Dorsa Sadigh, Yanan Sui Contact : szhang21@mit.edu Links: Paper | Video | Website Keywords : imitation learning, learning from demonstration, learning from suboptimal demonstrations Explaining heterogeneity in medial entorhinal cortex with task-driven neural networks Authors : Aran Nayebi, Alexander Attinger, Malcolm G. Campbell, Kiah Hardcastle, Isabel I.C. Low, Caitlin S. Mallory, Gabriel C. Mel, Ben Sorscher, Alex H. Williams, Surya Ganguli, Lisa M. Giocomo, Daniel L.K. Yamins Contact : anayebi@stanford.edu Award nominations: Spotlight Presentation Links: Paper | Website Keywords : neural coding, medial entorhinal cortex, grid cells, biologically-inspired navigation, path integration, recurrent neural networks On the theory of reinforcement learning with once-per-episode feedback Authors : Niladri Chatterji, Aldo Pacchiano, Peter Bartlett, Michael Jordan Contact : niladri@cs.stanford.edu Keywords : theoretical reinforcement learning, binary rewards, non-markovian rewards HyperSPNs: Compact and Expressive Probabilistic Circuits Authors : Andy Shih, Dorsa Sadigh, Stefano Ermon Contact : andyshih@stanford.edu Links: Paper | Video | Website Keywords : generative models, tractable probabilistic models, sum product networks, probabilistic circuits COMBO: Conservative Offline Model-Based Policy Optimization Authors : Tianhe Yu*, Aviral Kumar*, Rafael Rafailov, Aravind Rajeswaran, Sergey Levine, Chelsea Finn Contact : tianheyu@cs.stanford.edu Links: Paper Keywords : offline reinforcement learning, model-based reinforcement learning, deep reinforcement learning Conservative Data Sharing for Multi-Task Offline Reinforcement Learning Authors : Tianhe Yu*, Aviral Kumar*, Yevgen Chebotar, Karol Hausman, Sergey Levine, Chelsea Finn Contact : tianheyu@cs.stanford.edu Links: Paper Keywords : offline reinforcement learning, multi-task reinforcement learning, deep reinforcement learning Autonomous Reinforcement Learning via Subgoal Curricula Authors : Archit Sharma, Abhishek Gupta, Sergey Levine, Karol Hausman, Chelsea Finn Contact : architsh@stanford.edu Links: Paper | Website Keywords : reinforcement learning, curriculum, autonomous learning, reset-free reinforcement learning Lossy Compression for Lossless Prediction Authors : Yann Dubois, Benjamin Bloem-Reddy, Karen Ullrich Chris J. Maddison Contact : yanndubs@stanford.edu Award nominations: Spotlight Presentation Links: Paper | Video | Website Keywords : compression, invariances, information theory, machine learning, self-supervised learning Capturing implicit hierarchical structure in 3D biomedical images with self-supervised hyperbolic representations Authors : Joy Hsu, Jeffrey Gu, Gong-Her Wu, Wah Chiu, Serena Yeung Contact : joycj@stanford.edu Links: Paper Keywords : hyperbolic representations, hierarchical structure, biomedical Estimating High Order Gradients of the Data Distribution by Denoising Authors : Chenlin Meng, Yang Song, Wenzhe Li, Stefano Ermon Contact : chenlin@stanford.edu Keywords : score matching, langevin dynamics, denoising, generative modeling Universal Off-Policy Evaluation Authors : Yash Chandak, Scott Niekum, Bruno Castro da Silva, Erik Learned-Miller, Emma Brunskill, Philip Thomas Contact : ychandak@cs.umass.edu Links: Paper | Website Keywords : metrics, risk, distribution, cdf, off-policy evaluation, ope, reinforcement learning, counterfactuals, high-confidence bounds, confidence intervals Evidential Softmax for Sparse Multimodal Distributions in Deep Generative Models Authors : Phil Chen, Masha Itkina, Ransalu Senanayake, Mykel J. Kochenderfer Contact : philhc@stanford.edu Links: Paper Keywords : deep learning or neural networks, sparsity and feature selection, variational inference, (application) natural language and text processing Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss Authors : Jeff Z. HaoChen, Colin Wei, Adrien Gaidon, Tengyu Ma Contact : jhaochen@stanford.edu Links: Paper Keywords : deep learning theory, unsupervised learning theory, representation learning theory Provable Model-based Nonlinear Bandit and Reinforcement Learning: Shelve Optimism, Embrace Virtual Curvature Authors : Kefan Dong, Jiaqi Yang, Tengyu Ma Contact : kefandong@stanford.edu Links: Paper | Video Keywords : nonlinear bandits, online learning, deep reinforcement learning theory, sequential rademacher complexity Decrypting Cryptic Crosswords: Semantically Complex Wordplay Puzzles as a Target for NLP Authors : Joshua Rozner, Christopher Potts, Kyle Mahowald Contact : rozner@stanford.edu Links: Paper | Website Keywords : compositionality in language, curriculum learning, meta-linguistics, systematicity, generalization Design of Experiments for Stochastic Contextual Linear Bandits Authors : Andrea Zanette*, Kefan Dong*, Jonathan Lee*, Emma Brunskill Contact : zanette@berkeley.edu Links: Paper Keywords : linear bandits, design of experiments Provable Benefits of Actor-Critic Methods for Offline Reinforcement Learning Authors : Andrea Zanette, Martin J. Wainwright, Emma Brunskill Contact : zanette@berkeley.edu Links: Paper Keywords : offline rl, mirror descent, bellman closure A Topological Perspective on Causal Inference Authors : Duligur Ibeling, Thomas Icard Contact : icard@stanford.edu Links: Paper Keywords : causal inference, topological learning theory Adversarial Training Helps Transfer Learning via Better Representations Authors : Zhun Deng, Linjun Zhang, Kailas Vodrahalli, Kenji Kawaguchi, James Zou Contact : jamesyzou@gmail.com Links: Paper Keywords : transfer learning, adversarial training Widening the Pipeline in Human-Guided Reinforcement Learning with Explanation and Context-Aware Data Augmentation Authors : Lin Guan,Mudit Verma,Sihang Guo,Ruohan Zhang,Subbarao Kambhampati Contact : zharu@stanford.edu Award nominations: Spotlight Links: Paper | Website Keywords : human-in-the-loop reinforcement learning, evaluative feedback, saliency map, visual explanation Machine versus Human Attention in Deep Reinforcement Learning Tasks Authors : Sihang Guo, Ruohan Zhang, Bo Liu, Yifeng Zhu, Dana Ballard, Mary Hayhoe, Peter Stone Contact : zharu@stanford.edu Links: Paper Keywords : deep reinforcement learning, interpretability, attention, eye tracking Play to Grade: Testing Coding Games as Classifying Markov Decision Process Authors : Allen Nie, Emma Brunskill, Chris Piech Contact : anie@stanford.edu Links: Paper | Website Keywords : reinforcement learning, computational education, collaborative training, markov decision process The Value of Information When Deciding What to Learn Authors : Dilip Arumugam, Benjamin Van Roy Contact : dilip@cs.stanford.edu Links: Paper Keywords : exploration, information theory, multi-armed bandits, reinforcement learning Diversity Matters When Learning From Ensembles Authors : Giung Nam*, Jongmin Yoon*, Yoonho Lee, Juho Lee Contact : yoonho@cs.stanford.edu Links: Paper | Website Keywords : deep ensembles, knowledge distillation, calibration, output diversified sampling, batchensemble Reinforcement Learning with State Observation Costs in Action-Contingent Noiselessly Observable Markov Decision Processes Authors : HyunJi Nam, Scott Fleming, Emma Brunskill Contact : scottyf@stanford.edu Links: Paper | Website Keywords : reinforcement learning, observation cost, markov decision process, mdp, partially observable markov decision process, pomdp, probably approximately correct, pac, healthcare, health care Meta-learning with an Adaptive Task Scheduler Authors : Huaxiu Yao, Yu Wang, Ying Wei, Peilin Zhao, Mehrdad Mahdavi, Defu Lian, Chelsea Finn Contact : huaxiu@cs.stanford.edu Links: Paper Keywords : adaptive task scheduler, meta-learning, sampling Spatial-Temporal Super-Resolution of Satellite Imagery via Conditional Pixel Synthesis Authors : Yutong He, Dingjie Wang, Nicholas Lai, William Zhang, Chenlin Meng, Marshall Burke, David B. Lobell, Stefano Ermon Contact : kellyyhe@stanford.edu Links: Paper | Video | Website Keywords : remote sensing, super-resolution, generative models Scatterbrain: Unifying Sparse and Low-rank Attention Authors : Beidi Chen*, Tri Dao*, Eric Winsor, Zhao Song, Atri Rudra, Christopher Ré. Contact : trid@stanford.edu Links: Paper Keywords : efficient attention, sparse, low-rank BCD Nets: Scalable Variational Approaches for Bayesian Causal Discovery Authors : Chris Cundy, Aditya Grover, Stefano Ermon Contact : cundy@stanford.edu Keywords : causal inference, variational inference Calibrating Predictions to Decisions: A Novel Approach to Multi-Class Calibration Authors : Shengjia Zhao, Michael P Kim, Roshni Sahoo, Tengyu Ma, Stefano Ermon Contact : sjzhao@stanford.edu Links: Paper Keywords : calibration, decision making under uncertainty Beyond Pinball Loss: Quantile Methods for Calibrated Uncertainty Quantification Authors : Youngseog Chung, Willie Neiswanger, Ian Char, Jeff Schneider Contact : youngsec@andrew.cmu.edu, neiswanger@cs.stanford.edu Links: Paper | Website Keywords : uncertainty quantification, uq, quantile regression, pinball loss Causal Abstractions of Neural Networks Authors : Atticus Geiger*, Hanson Lu*, Thomas Icard, Christopher Potts Contact : atticusg@stanford.edu Links: Paper Keywords : interpretability, analysis, nlp, causality Generalized Shape Metrics on Neural Representations Authors : Alex H Williams, Erin Kunz, Simon Kornblith, Scott Linderman Contact : alex.h.willia@gmail.com Keywords : representational similarity analysis, neural representations, shape analysis, metric space D2C: Diffusion-Denoising Models for Few-shot Conditional Generation Authors : Abhishek Sinha*, Jiaming Song*, Chenlin Meng, Stefano Ermon Contact : tsong@cs.stanford.edu Links: Paper | Website Keywords : generative modeling, contrastive learning, conditional generation Combiner: Full Attention Transformer with Sparse COmputation Cost Authors : Hongyu Ren, Hanjun Dai, Zihang Dai, Mengjiao Yang, Jure Leskovec, Dale Schuurmans, Bo Dai Contact : hyren@cs.stanford.edu Links: Paper Keywords : efficient transformer Maximum Likelihood Training of Score-Based Diffusion Models Authors : Yang Song, Conor Durkan, Iain Murray, Stefano Ermon Contact : yangsong@cs.stanford.edu Award nominations: Spotlight presentation Links: Paper Keywords : score-based generative models, denoising score matching, diffusion models, maximum likelihood training Contrastive Reinforcement Learning of Symbolic Reasoning Domains Authors : Gabriel Poesia, WenXin Dong, Noah Goodman Contact : poesia@stanford.edu Keywords : reinforcement learning, education, contrastive learning, symbolic reasoning Equivariant Manifold Flows Authors : Isay Katsman, Aaron Lou, Derek Lim, Qingxuan Jiang, Ser Nam Lim, Christopher M. De Sa Contact : aaronlou@stanford.edu Links: Paper | Website Keywords : manifold, normalizing flow, equivariant, invariant Lower Bounds on Metropolized Sampling Methods for Well-Conditioned Distributions Authors : Yin Tat Lee, Ruoqi Shen, Kevin Tian Contact : kjtian@stanford.edu Award nominations: Oral presentation Links: Paper | Video Keywords : sampling, lower bounds, langevin dynamics, hamiltonian monte carlo List-Decodable Mean Estimation in Nearly-PCA Time Authors : Ilias Diakonikolas, Daniel M. Kane, Daniel Kongsgaard, Jerry Li, Kevin Tian Contact : kjtian@stanford.edu Award nominations: Spotlight presentation Links: Paper Keywords : robust statistics, semidefinite programming, mixture models Robust Regression Revisited: Acceleration and Improved Estimation Rates Authors : Arun Jambulapati, Jerry Li, Tselil Schramm, Kevin Tian Contact : kjtian@stanford.edu Links: Paper Keywords : robust statistics, regression, generalized linear models, acceleration, sum of squares methods Learning with User-Level Privacy Authors : Daniel Levy*, Ziteng Sun*, Kareem Amin, Satyen Kale, Alex Kulesza, Mehryar Mohri, Ananda Theertha Suresh Contact : danilevy@stanford.edu Links: Paper Keywords : differential privacy user-level Adapting to Function Difficulty and Growth Conditions in Private Optimization Authors : Hilal Asi*, Daniel Levy*, John C. Duchi Contact : asi@stanford.edu Links: Paper Keywords : differential privacy adaptivity optimization Imitation with Neural Density Models Authors : Kuno Kim, Akshat Jindal, Yang Song, Jiaming Song, Yanan Sui, Stefano Ermon Contact : khkim@cs.stanford.edu Links: Paper Keywords : rl; imitation learning; density estimation Why Do Pretrained Language Models Help in Downstream Tasks? An Analysis of Head and Prompt Tuning Authors : Colin Wei, Sang Michael Xie, Tengyu Ma Contact : colinwei@stanford.edu Links: Paper Keywords : nlp pretraining, theoretical analysis Safe Reinforcement Learning by Imagining the Near Future Authors : Garrett Thomas, Yuping Luo, Tengyu Ma Contact : gwthomas@stanford.edu Links: Paper Keywords : safe exploration, model-based rl Pseudo-Spherical Contrastive Divergence Authors : Lantao Yu, Jiaming Song, Yang Song, Stefano Ermon Contact : lantaoyu@cs.stanford.edu Links: Paper Keywords : deep generative models, energy-based models, proper scoring rules IQ-Learn: Inverse soft-Q Learning for Imitation Authors : Divyansh Garg, Shuvam Chakraborty, Chris Cundy, Jiaming Song, Stefano Ermon Contact : divgarg@stanford.edu Award nominations: Spotlight Links: Paper | Website Keywords : reinforcement learning, imitation learning, inverse reinforcement learning, statistical learning, energy-based models Intrinsic Dimension, Persistent Homology and Generalization in Neural Networks Authors : Tolga Birdal ~Tolga_Birdal3 , Aaron Lou, Leonidas Guibas, Umut Simsekli Contact : aaronlou@stanford.edu Links: Paper | Website Keywords : generalization, persistent homology, intrinsic dimension, deep networks Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval Authors : Omar Khattab, Christopher Potts, Matei Zaharia Contact : okhattab@stanford.edu Award nominations: Spotlight paper Links: Paper | Blog Post Keywords : neural retrieval, multi-hop question answering, claim verification, reasoning, colbert Datasets and Benchmarks Track ReaSCAN: Compositional Reasoning in Language Grounding | Website by Zhengxuan Wu*, Elisa Kreiss*, Desmond Ong, Christopher Potts ATOM3D: Tasks on Molecules in Three Dimensions | Website by Raphael J.L. Townshend, Martin Vögele, Patricia Suriana, Alexander Derry, Alexander S. Powers, Yianni Laloudakis, Sidhika Balachandar, Bowen Jing, Brandon Anderson, Stephan Eismann, Risi Kondor, Russ B. Altman, Ron O. Dror Dynamic Environments with Deformable Objects | Video | Website by Rika Antonova, Peiyang Shi, Hang Yin, Zehang Weng, Danica Kragic Personalized Benchmarking with the Ludwig Benchmarking Toolkit | Website by Avanika Narayan, Piero Molino, Karan Goel, Willie Neiswanger, Christopher Ré SKM-TEA: A Dataset for Accelerated MRI Reconstruction with Dense Image Labels for Quantitative Clinical Evaluation | Website by Arjun D Desai, Andrew M Schmidt, Elka B Rubin, Christopher M Sandino, Marianne S Black, Valentina Mazzoli, Kathryn J Stevens, Robert Boutin, Christopher Ré, Garry E Gold, Brian A Hargreaves, Akshay S Chaudhari Are We Learning Yet? A Meta Review of Evaluation Failures Across Machine Learning by Thomas Liao, Rohan Taori, Inioluwa Deborah Raji, Ludwig Schmidt DABS: a Domain-Agnostic Benchmark for Self-Supervised Learning | Website by Alex Tamkin, Vincent Liu, Rongfei Lu, Daniel Fein, Colin Schultz, Noah Goodman SustainBench: Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning | Video | Website by Christopher Yeh, Chenlin Meng, Sherrie Wang, Anne Driscoll, Erik Rozi, Patrick Liu, Jihyeon Lee, Marshall Burke, David Lobell, Stefano Ermon OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphs | Website by Weihua Hu Workshops This year, multiple members of the SAIL community are also involved in great workshops that will take place on Dec 13-14. We hope you’ll check them out! Machine Learning for Structural Biology Workshop (Dec 13) Organizers : Namrata Anand, Bonnie Berger, Wouter Boomsma, Erika DeBenedictis, Stephan Eismann, John Ingraham, Sergey Ovchinnikov, Roshan Rao, Raphael Townshend and Ellen Zhong Controllable Generative Modeling in Language and Vision (CtrlGen Workshop) (Dec 13) Organizers : Steven Y. Feng, Drew A. Hudson, Anusha Balakrishnan, Varun Gangal, Dongyeop Kang, Tatsunori Hashimoto and Joel Tetreault DistShift Workshop (Dec 13) Organizers : Shiori Sagawa, Pang Wei Koh, Fanny Yang, Hongseok Namkoong, Jiashi Feng, Kate Saenko, Percy Liang, Sarah Bird and Sergey Levine Data-centric AI Workshop (Dec 14) Organizers : Andrew Ng, Lora Aroyo, Cody Coleman, Greg Diamos, Vijay Janapa Reddi, Joaquin Vanschoren,Carole-Jean Wu and Sharon Zhou Physical Reasoning and Inductive Biases for the Real World Workshop (Dec 14) Organizers : Krishna Murthy Jatavallabhula, Rika Antonova, Kevin Smith, Hsiao-Yu (Fish) Tung, Florian Shkurti, Jeannette Bohg and Josh Tenenbaum Workshop Papers How Does Contrastive Pre-training Connect Disparate Domains? by Kendrick Shen*, Robbie Jones*, Ananya Kumar*, Sang Michael Xie*, Percy Liang ( DistShift Workshop ) Optimal Representations for Covariate Shifts by Yann Dubois, Yangjun