Grid'5000 experiment

Jump to: navigation, search

Rollout Allocation Strategies for Classification-based Policy Iteration (Programming)

Conducted by

Victor GABILLON

Description

Reinforcement Learning: In classification-based policy iteration algorithms: 1) the value function over a finite number of states, called the rollout set, and the actions in the action space, are estimated through rollouts in order to 2) learn an improved policy through a classifier. The choice of rollout allocation strategies over states and actions have significant impact on the performance and computation time of this class of algorithms. In this paper, we present new strategies to allocate the available budget (number of rollouts) at each iteration of the algorithm over states, actions, and state-action pairs. Our empirical results indicate that, for a fixed budget, and in comparison to existing methods, using the proposed strategies improves the accuracy of the training set.

Status

in progress

Resources

    Tools used

    No information

    Results

    Not yet

    Shared by: Victor GABILLON
    Last update: 2010-06-10 14:46:52
    Experiment #665

    Personal tools
    Namespaces

    Variants
    Views
    Actions
    Public Portal
    Users Portal
    Admin portal
    Wiki special pages
    Toolbox