n South African Computer Journal - Upper bounds on the performance of discretisation in reinforcement learning : research article
|Article Title||Upper bounds on the performance of discretisation in reinforcement learning : research article|
|© Publisher:||South African Computer Society (SAICSIT)|
|Journal||South African Computer Journal|
|Affiliations||1 University of the Witwatersrand|
|Publication Date||Dec 2015|
|Pages||24 - 31|
|Keyword(s)||Average case analysis, Performance bounds, Reinforcement learning and Tile coding|
Reinforcement learning is a machine learning framework whereby an agent learns to perform a task by maximising its total reward received for selecting actions in each state. The policy mapping states to actions that the agent learns is either represented explicitly, or implicitly through a value function. It is common in reinforcement learning to discretise a continuous state space using tile coding or binary features. We prove an upper bound on the performance of discretisation for direct policy representation or value function approximation.
Article metrics loading...