Continuous Control with Action Quantization from Demonstrations
By Robert Dadashi, L'eonard Hussenot, Damien Vincent, Sertan Girgin, Anton Raichuk, Matthieu Geist, Olivier Pietquin. 2021
33 citations recorded by the source corpus.
In this paper, we propose a novel Reinforcement Learning (RL) framework for problems with continuous action spaces: Action Quantization from Demonstrations (AQuaDem). The proposed approach consists in learning a discretization of continuous action spaces from human demonstrations. This discretization returns a set of plausible actions (in light of the demonstrations) for each input state, thus capturing the priors of the demonstrator and their multimodal behavior. By discretizing the action space, any discrete action deep RL technique can be readily applied to the continuous control problem. Experiments show that the proposed approach outperforms state-of-the-art methods such as SAC in the RL setup, and GAIL in the Imitation Learning setup. We provide a website with interactive videos: https://google-research.github.io/aquadem/ and make the code available: https://github.com/google-research/google-research/tree/master/aquadem.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Sergey Levine
- Ziyun Wang
- T. Paine
- Yoshua Bengio
- Caglar Gulcehre
- Bobak Shahriari
- Matthew W. Hoffman
- Ruslan Salakhutdinov
Citations in the corpus
- Acme: A Research Framework for Distributed Reinforcement Learning (2020, background): “Learning from Demonstrations In the Learning from Demonstrations (LfD) setting (Schaal, 1997; Hester et al., 2018; Vecerik et al., 2017a; Dadashi et al., 2022), similarly to the standard online RL setting, the goal is to learn a policy through trial and error by interacting with an environment, whic”