Soft Actor-Critic for Discrete Action Settings
By Petros Christodoulou. arXiv.org, 2019
431 citations recorded by the source corpus.
Soft Actor-Critic is a state-of-the-art reinforcement learning algorithm for continuous action settings that is not applicable to discrete action settings. Many important settings involve discrete actions, however, and so here we derive an alternative version of the Soft Actor-Critic algorithm that is applicable to discrete action settings. We then show that, even without any hyperparameter tuning, it is competitive with the tuned model-free state-of-the-art on a selection of games from the Atari suite.
Subfields: Artificial Intelligence, Sociology and Political Science
View Paper: https://www.semanticscholar.org/paper/0a0866ec7180bbf87b1c87ed48bd4fa00574b814
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sergey Levine
- Matthew Riemer
- Ofir Nachum
- G. Tesauro
- Miao Liu
- Jonathan P. How
- Doina Precup
- Pieter Abbeel
In Sociology and Political Science
- Michael Everett
- Doina Precup
- Jakob N. Foerster
- Dong-Ki Kim
- G. Tesauro
- Jonathan P. How
- Chuangchuang Sun
- Elaine Lau
Citations in the corpus
- Influencing Long-Term Behavior in Multiagent Reinforcement Learning (2022, positive): “The inferred information becomes the input to the reinforcement learning module, which extends the policy gradient theorem in Equation (10) and learns active average reward policies sample efficiently by building on the multiagent soft actor-critic (MASAC) framework [14, 27, 31].”
- Policy Gradients Incorporating the Future (2021, methodology): “We evaluate PGIF-versions of SAC and PPO to vanilla discreteSAC (Haarnoja et al., 2018; Christodoulou, 2019) as well as a PPO-LSTM (Schulman et al.”