Reinforcement learning by directly learning an advantage function

Yunhao Tang; Rémi Munos; Mark Daniel Rowland; Michal Valko

Reinforcement learning by directly learning an advantage function

Yunhao Tang, Rémi Munos, Mark Daniel Rowland, Michal Valko

2024 · US Patent App. 18/424,520 (2024)

Abstract

A system and method of controlling an agent to take actions in an environment to perform a task by directly learning an advantage function through reinforcement learning.

PDF · US Patent App. 18/424,520

Michal Valko : Paper

Reinforcement learning by directly learning an advantage function

Abstract