[1908.09357] Dynamics-aware Embeddings - arXiv.org
Aug 25, 2019 · We propose a forward prediction objective for simultaneously learning embeddings of states and action sequences. These embeddings capture the structure of the environment's dynamics, enabling efficient policy learning.
- Author: William Whitney, Rajat Agarwal, Kyunghyun Cho, Abhinav Gupta
- Cite as: arXiv:1908.09357 [cs.LG]
- Publish Year: 2019
- Comments: Published at ICLR 2020
See results only from arxiv.orgar5iv
These embeddings capture the structure of the environment’s dynamics, enabling …
willwhitney/dynamics-aware-embeddings - GitHub
- This repository contains the code for "Dynamics-aware Embeddings" by William Whitney, Rajat …
arXiv link
- This repository contains the code for "Dynamics-aware Embeddings" by William Whitney, Rajat …
Dynamics-aware Embeddings | Facebook AI Research - AI at Meta
Apr 25, 2020 · We propose a forward prediction objective for simultaneously learning embeddings of states and action sequences. These embeddings capture the structure of the environment’s dynamics, enabling efficient policy learning.
Dynamics-aware Embeddings - Papers With Code
We propose a forward prediction objective for simultaneously learning embeddings of states and action sequences. These embeddings capture the structure of the environment's dynamics, enabling efficient policy learning.
DYNAMICS-AWARE EMBEDDINGS — NYU Scholars
We propose a forward prediction objective for simultaneously learning embeddings of states and action sequences. These embeddings capture the structure of the environment's dynamics, enabling efficient policy learning.
- Country/Territory: Ethiopia
- State: Published-2020
By combining state and action embeddings, we achieve efficient learning of high-quality policies on goal-conditioned continuous control from pixel observations in only 1-2 million environment steps.
[1908.09357] Dynamics-aware Embeddings - ar5iv
These embeddings capture the structure of the environment’s dynamics, enabling efficient policy learning. We demonstrate that our action embeddings alone improve the sample efficiency and peak performance of model-free RL on control from low-dimensional states.
Dynamics-aware Embeddings - DeepAI
Aug 25, 2019 · We propose a forward prediction objective for simultaneously learning embeddings of states and actions. These embeddings capture the structure of the environment's dynamics, enabling efficient policy learning.
ICLR Poster Dynamics-Aware Embeddings
We propose a forward prediction objective for simultaneously learning embeddings of states and actions. These embeddings capture the structure of the environment's dynamics, enabling efficient policy learning.
[PDF] Dynamics-aware Embeddings - Semantic Scholar
Aug 25, 2019 · We propose a forward prediction objective for simultaneously learning embeddings of states and action sequences. These embeddings capture the structure of the environment's dynamics, enabling efficient policy learning.