FeUdal Networks for Hierarchical Reinforcement Learning
This addresses the problem of long-term credit assignment in reinforcement learning for AI agents, offering a novel hierarchical approach that is not incremental but builds on prior feudal RL concepts.
The paper tackles hierarchical reinforcement learning by introducing FeUdal Networks (FuNs), which decouple learning across Manager and Worker modules to handle long-term credit assignment, resulting in dramatic performance improvements over a strong baseline on tasks like ATARI and DeepMind Lab environments.
We introduce FeUdal Networks (FuNs): a novel architecture for hierarchical reinforcement learning. Our approach is inspired by the feudal reinforcement learning proposal of Dayan and Hinton, and gains power and efficacy by decoupling end-to-end learning across multiple levels -- allowing it to utilise different resolutions of time. Our framework employs a Manager module and a Worker module. The Manager operates at a lower temporal resolution and sets abstract goals which are conveyed to and enacted by the Worker. The Worker generates primitive actions at every tick of the environment. The decoupled structure of FuN conveys several benefits -- in addition to facilitating very long timescale credit assignment it also encourages the emergence of sub-policies associated with different goals set by the Manager. These properties allow FuN to dramatically outperform a strong baseline agent on tasks that involve long-term credit assignment or memorisation. We demonstrate the performance of our proposed system on a range of tasks from the ATARI suite and also from a 3D DeepMind Lab environment.