Improvisation through Physical Understanding: Using Novel Objects as Tools with Visual Foresight
Comments Videos available at https://sites.google.com/view/gvf-tool
作者
Chelsea Finn
Robotics / Machine Learning
Unsupervised Learning via Meta-Learning
Comments ICLR 2019 camera-ready. 24 pages, 2 figures, links to code available at https://sites.google.com/view/unsupervised-via-meta
Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables
Manipulation by Feel: Touch-Based Control with Deep Predictive Models
Comments Accepted to ICRA 2019
NoRML: No-Reward Meta Learning
Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learning
Comments First 2 authors contributed equally. Website: https://sites.google.com/berkeley.edu/metaadaptivecontrol
Unsupervised Visuomotor Control through Distributional Planning Networks
Comments Videos available at https://sites.google.com/view/dpn-public/
Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RL
Comments Project website: https://sites.google.com/berkeley.edu/onlineviameta
Reasoning About Physical Interactions with Object-Oriented Prediction and Planning
Comments ICLR 2019, project page: https://people.eecs.berkeley.edu/~janner/o2p2/
Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control
One-Shot Hierarchical Imitation Learning of Compound Visuomotor Tasks
Comments Video results available at https://sites.google.com/view/one-shot-hil
Robustness via Retrying: Closed-Loop Robotic Manipulation with Self-Supervised Learning
Comments accepted at the Conference on Robot Learning (CoRL) 2018
Few-Shot Goal Inference for Visuomotor Learning and Planning
Comments Videos available at https://sites.google.com/view/few-shot-goals
Stochastic Adversarial Video Prediction
Comments Website: https://alexlee-gk.github.io/video_prediction/
Universal Planning Networks
Comments Videos available at https://sites.google.com/view/upn-public/home
Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods
Comments 8 pages
Stochastic Variational Video Prediction
Meta-Learning and Universality: Deep Representations and Gradient Descent can Approximate any Learning Algorithm
Comments ICLR 2018
One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning
Comments First two authors contributed equally. Video available at https://sites.google.com/view/daml
Recasting Gradient-Based Meta-Learning as Hierarchical Bayes
Self-Supervised Visual Planning with Temporal Skip Connections
Comments accepted at the Conference on Robot Learning (CoRL) 2017
One-Shot Visual Imitation Learning via Meta-Learning
Comments Conference on Robot Learning, 2017 (to appear). First two authors contributed equally. Video available at https://sites.google.com/view/one-shot-imitation
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
Comments ICML 2017. Code at https://github.com/cbfinn/maml, Videos of RL results at https://sites.google.com/view/maml, Blog post at http://bair.berkeley.edu/blog/2017/07/18/learning-to-learn/
Adapting Deep Visuomotor Representations with Weak Pairwise Constraints
Deep Visual Foresight for Planning Robot Motion
Comments ICRA 2017. Supplementary video: https://sites.google.com/site/robotforesight/
Generalizing Skills with Semi-Supervised Reinforcement Learning
Comments ICLR 2017
Active One-shot Learning
Comments NIPS 2016, Deep Reinforcement Learning Workshop, Barcelona, Spain. See https://cs.stanford.edu/~woodward/ for the poster and a short video description of the paper
A Connection between Generative Adversarial Networks, Inverse Reinforcement Learning, and Energy-Based Models
Comments NIPS 2016 Workshop on Adversarial Training. First two authors contributed equally
Unsupervised Learning for Physical Interaction through Video Prediction
Comments To appear in NIPS '16; Video results, code, and data available at: http://www.sites.google.com/site/robotprediction
