Continual Learning of Control Primitives: Skill Discovery via Reset-Games
Comments To appear at NeurIPS 2020
作者
Chelsea Finn
Robotics / Machine Learning
Learning to be Safe: Deep RL with a Safety Critic
Comments In submission, 16 pages (including appendix)
Guided Meta-Policy Search
Comments Published at Neurips 2019. Website : https://sites.google.com/berkeley.edu/guided-metapolicy-search
Continuous Meta-Learning without Tasks
Comments NeurIPS 2020
Goal-Aware Prediction: Learning to Model What Matters
Never Stop Learning: The Effectiveness of Fine-Tuning in Robotic Reinforcement Learning
Comments 8.5 pages, 9 figures. See video overview and experiments at https://youtu.be/pPDVewcSpdc and project website at https://ryanjulian.me/continual-fine-tuning
Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning
Comments Published as a conference paper at International Conference on Intelligent Robots and Systems (IROS) 2020. See http://youtu.be/_QPMCDdFC3E for associated video file, http://github.com/google-research/google-research/tree/master/es_maml for associated code, and https://ai.googleblog.com/2020/04/exploring-evolutionary-meta-learning-in.html for the corresponding Google AI Blog post
Deep Reinforcement Learning amidst Lifelong Non-Stationarity
Comments supplementary website at https://sites.google.com/stanford.edu/lilac/
Meta-Reinforcement Learning Robust to Distributional Shift via Model Identification and Experience Relabeling
Time Reversal as Self-Supervision
Comments 7 pages, 10 figures
Journal ref International Conference on Robotics and Automation, 2020
Entity Abstraction in Visual Model-Based Reinforcement Learning
Comments Accepted at CoRL 2019
Unsupervised Meta-Learning for Reinforcement Learning
Comments First two authors contributed equally
Meta-Learning without Memorization
Comments ICLR 2020
OmniTact: A Multi-Directional High Resolution Touch Sensor
Comments Accepted at International Conference on Robotics and Automation (ICRA) 2020
Scalable Multi-Task Imitation Learning with Autonomous Improvement
Comments Accepted to ICRA 2020. Supplementary material at https://sites.google.com/view/scalable-mili
VideoFlow: A Conditional Flow-Based Model for Stochastic Video Generation
Comments ICLR 2020 Camera-Ready. Previous title: VideoFlow: A Flow-Based Generative Model for Video
Watch, Try, Learn: Meta-Learning from Demonstrations and Reward
RoboNet: Large-Scale Multi-Robot Learning
Comments accepted at the Conference on Robot Learning (CoRL) 2019
Learning Predictive Models From Observation and Interaction
Unsupervised Curricula for Visual Meta-Reinforcement Learning
Comments NeurIPS 2019
Learning to Interactively Learn and Assist
Comments AAAI 2020. Video overview at https://youtu.be/8yBvDBuAPrw, paper website with videos and interactive game at http://interactive-learning.github.io/
Language as an Abstraction for Hierarchical Deep Reinforcement Learning
Comments Published in Neural Information Processing Systems (NeurIPS) 2019; Supplementary materials: https://sites.google.com/view/hal-demo
Meta-Inverse Reinforcement Learning with Probabilistic Context Variables
Comments NeurIPS 2019
Probabilistic Model-Agnostic Meta-Learning
Comments NeurIPS 2018. First two authors contributed equally. Supplementary results available at https://sites.google.com/view/probabilistic-maml/
Learning a Prior over Intent via Meta-Inverse Reinforcement Learning
Hierarchical Foresight: Self-Supervised Learning of Long-Horizon Tasks via Visual Subgoal Generation
Comments 16 pages, 9 figures
Meta-Learning with Implicit Gradients
Comments NeurIPS 2019. First two authors contributed equally
Online Meta-Learning
Comments ICML 2019. The first two authors contributed equally. Expanded Appendix
Training an Interactive Helper
Comments The paper "Learning to Interactively Learn and Assist" (LILA), at arXiv:1906.10187, supersedes this paper. This preliminary workshop paper appeared in the Emergent Communication Workshop and Workshop on Learning by Instruction at NeurIPS 2018
End-to-End Robotic Reinforcement Learning without Reward Engineering
Comments Accepted to RSS 2019. 14 pages and 13 figures including references and appendix. Website: https://sites.google.com/view/reward-learning-rl/home
