Reuse of Neural Modules for General Video Game Playing
This addresses the challenge of efficient learning in sequential decision-making domains for AI agents, though it appears incremental as it builds on prior neural transfer methods.
The paper tackles the problem of knowledge transfer in reinforcement learning by introducing a domain-agnostic approach where neural networks reuse existing modules to improve performance in new tasks, demonstrating gains in complex Atari 2600 games and predicting transfer success based on game dynamics.
A general approach to knowledge transfer is introduced in which an agent controlled by a neural network adapts how it reuses existing networks as it learns in a new domain. Networks trained for a new domain can improve their performance by routing activation selectively through previously learned neural structure, regardless of how or for what it was learned. A neuroevolution implementation of this approach is presented with application to high-dimensional sequential decision-making domains. This approach is more general than previous approaches to neural transfer for reinforcement learning. It is domain-agnostic and requires no prior assumptions about the nature of task relatedness or mappings. The method is analyzed in a stochastic version of the Arcade Learning Environment, demonstrating that it improves performance in some of the more complex Atari 2600 games, and that the success of transfer can be predicted based on a high-level characterization of game dynamics.