Multicopy Reinforcement Learning Agents
This addresses a novel multi-agent problem for reinforcement learning, potentially enhancing efficiency in tasks with environmental noise, though it appears incremental as it builds on existing multi-agent concepts.
The paper tackles the problem of a single agent making multiple identical copies of itself to improve performance in noisy environments where tasks may be unachievable by a single copy, proposing a learning algorithm that efficiently balances the benefits and costs of additional copies.
This paper examines a novel type of multi-agent problem, in which an agent makes multiple identical copies of itself in order to achieve a single agent task better or more efficiently. This strategy improves performance if the environment is noisy and the task is sometimes unachievable by a single agent copy. We propose a learning algorithm for this multicopy problem which takes advantage of the structure of the value function to efficiently learn how to balance the advantages and costs of adding additional copies.