Fast Model Editing at Scale
Comments ICLR 2022. View implementation and additional project info at https://sites.google.com/view/mend-editing
作者
Chelsea Finn
Robotics / Machine Learning
Play it by Ear: Learning Skills amidst Occlusion through Audio-Visual Imitation Learning
Journal ref Robotics Science and Systems (RSS) 2022
A State-Distribution Matching Approach to Non-Episodic Reinforcement Learning
Extending the WILDS Benchmark for Unsupervised Adaptation
Training and Evaluation of Deep Policies using Reinforcement Learning and Generative Models
Comments arXiv admin note: substantial text overlap with arXiv:2007.13134
Lifelong Robotic Reinforcement Learning by Retaining Experiences
Comments Supplementary website at https://sites.google.com/view/retain-experience/
Vision-Based Manipulators Need to Also See from Their Hands
Comments First two authors contributed equally. ICLR 2022 (oral) camera-ready. 30 pages, 20 figures. Project website: https://sites.google.com/view/seeing-from-hands
Do Deep Networks Transfer Invariances Across Classes?
Latent-Variable Advantage-Weighted Policy Optimization for Offline RL
Meta-Learning with Fewer Tasks through Task Interpolation
Comments Accepted by ICLR 2022 (Oral)
Policy Architectures for Compositional Generalization in Control
Robust Policy Learning over Multiple Uncertainty Sets
Comments Project webpage at https://sites.google.com/view/sirsa-public/home
Fully Online Meta-Learning Without Task Boundaries
BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning
Comments CoRL 2021, 23 pages
Journal ref Conference on Robot Learning (pp. 991-1002). 2022 Jan 11
COMBO: Conservative Offline Model-Based Policy Optimization
Comments NeurIPS 2021
CoMPS: Continual Meta Policy Search
Comments 23 pages, under review
Information is Power: Intrinsic Control via Information Capture
Comments NeurIPS 2021
MESA: Offline Meta-RL for Safe Adaptation and Fault Tolerance
Journal ref Workshop on Safe and Robust Control of Uncertain Systems at the 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Online
Noether Networks: Meta-Learning Useful Conserved Quantities
Comments Accepted to NeurIPS '21. The first two authors contributed equally
Adaptive Risk Minimization: Learning to Adapt to Domain Shift
Comments NeurIPS 2021 ; Project website: https://sites.google.com/view/adaptive-risk-minimization ; Code: https://github.com/henrikmarklund/arm
Decoupling Exploration and Exploitation for Meta-Reinforcement Learning without Sacrifices
Comments International Conference on Machine Learning (ICML), 2021
Reinforcement Learning with Videos: Combining Offline Observations with Interaction
Journal ref Conference on Robot Learning (2020)
Learning Language-Conditioned Robot Behavior from Offline Data and Crowd-Sourced Annotation
Comments Conference on Robot Learning (CoRL) 2021. 24 Pages, 18 Figures
Autonomous Reinforcement Learning via Subgoal Curricula
Meta-learning with an Adaptive Task Scheduler
Comments Accepted by NeurIPS 2021
Differentiable Annealed Importance Sampling and the Perils of Gradient Noise
Comments 23 pages, NeurIPS 2021
Efficiently Identifying Task Groupings for Multi-Task Learning
Comments In NeurIPS 2021 (spotlight). Code is available at https://github.com/google-research/google-research/tree/master/tag
ProtoTransformer: A Meta-Learning Approach to Providing Student Feedback
Comments 9 pages content; 6 pages supplement
Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasets
Just Train Twice: Improving Group Robustness without Training Group Information
Comments International Conference on Machine Learning (ICML), 2021
