Skip to content

State-Encompassing Training  #15

@SeverTopan

Description

@SeverTopan

RL processes such as Q-Learning are based mapping outputs to a set of input states I. If I is computable, would it be possible to train an agent by iterating though all the possible states systematically, as opposed to repeatedly running simulations and waiting for a sufficient number of states be reached through happenstance?

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions