T
Thanard Kurutach
Researcher at University of California, Berkeley
Publications - 18
Citations - 959
Thanard Kurutach is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Reinforcement learning & Deep learning. The author has an hindex of 10, co-authored 18 publications receiving 667 citations. Previous affiliations of Thanard Kurutach include Massachusetts Institute of Technology.
Papers
More filters
Posted Content
Model-Ensemble Trust-Region Policy Optimization
TL;DR: The authors proposed to use an ensemble of models to maintain the model uncertainty and regularize the learning process, which significantly reduces the sample complexity compared to model-free deep RL methods on challenging continuous control benchmark tasks.
Proceedings Article
Model-Ensemble Trust-Region Policy Optimization
TL;DR: This paper analyzes the behavior of vanilla model-based reinforcement learning methods when deep neural networks are used to learn both the model and the policy, and shows that the learned policy tends to exploit regions where insufficient data is available for the model to be learned, causing instability in training.
Proceedings Article
Learning plannable representations with causal InfoGAN
TL;DR: This work asks how to imagine goal-directed visual plans – a plausible sequence of observations that transition a dynamical system from its current configuration to a desired goal state, which can later be used as a reference trajectory for control.
Proceedings ArticleDOI
Learning to Manipulate Deformable Objects without Demonstrations
TL;DR: This paper proposes an iterative pick-place action space that encodes the conditional relationship between picking and placing on deformable objects and obtains an order of magnitude faster learning compared to independent action-spaces on a suite of deformable object manipulation tasks with visual RGB observations.
Proceedings Article
Learning Robotic Manipulation through Visual Planning and Acting.
TL;DR: This work learns to imagine goal-directed object manipulation directly from raw image data of self-supervised interaction of the robot with the object, and shows that separating the problem into visual planning and visual tracking control is more efficient and more interpretable than alternative data-driven approaches.