- Abstract: Prefrontal cortex (PFC) is a part of the brain which is responsible for behavior repertoire. Inspired by PFC functionality and connectivity, as well as human behavior formation process, we propose a novel modular architecture of neural networks with a Behavioral Module (BM) and corresponding end-to-end training strategy. This approach allows the efficient learning of behaviors and preferences representation. This property is particularly useful for user modeling (as for dialog agents) and recommendation tasks, as allows learning personalized representations of different user states. In the experiment with video games playing, the resultsshow that the proposed method allows separation of main task’s objectives andbehaviors between different BMs. The experiments also show network extendability through independent learning of new behavior patterns. Moreover, we demonstrate a strategy for an efficient transfer of newly learned BMs to unseen tasks.
- Keywords: Modular Networks, Reinforcement Learning, Task Separation, Representation Learning, Transfer Learning, Adversarial Transfer
- TL;DR: Extendable Modular Architecture is proposed for developing of variety of Agent Behaviors in DQN.