Convergent Actor-Critic Algorithms Under Off-Policy Training and Function Approximation
–arXiv.org Artificial Intelligence
We present the first class of policy-gradient algorithms that work with both state-value and policy function-approximation, and are guaranteed to converge under off-policy training. Our solution targets problems in reinforcement learning where the action representation adds to the-curse-of-dimensionality; that is, with continuous or large action sets, thus making it infeasible to estimate state-action value functions (Q functions). Using state-value functions helps to lift the curse and as a result naturally turn our policy-gradient solution into classical Actor-Critic architecture whose Actor uses state-value function for the update. Our algorithms, Gradient Actor-Critic and Emphatic Actor-Critic, are derived based on the exact gradient of averaged state-value function objective and thus are guaranteed to converge to its optimal solution, while maintaining all the desirable properties of classical Actor-Critic methods with no additional hyper-parameters. To our knowledge, this is the first time that convergent off-policy learning methods have been extended to classical Actor-Critic methods with function approximation.
arXiv.org Artificial Intelligence
Feb-21-2018
- Country:
- North America
- Canada > Alberta (0.14)
- United States
- Massachusetts > Hampshire County
- Amherst (0.14)
- California
- Santa Clara County > Palo Alto (0.04)
- San Diego County > San Diego (0.04)
- Massachusetts > Hampshire County
- Europe > United Kingdom
- Scotland > City of Edinburgh
- Edinburgh (0.04)
- England > Cambridgeshire
- Cambridge (0.04)
- Scotland > City of Edinburgh
- North America
- Genre:
- Research Report (0.64)
- Technology: