Near-Optimal Representation Learning for Hierarchical Reinforcement Learning

Code

• github.com/tensorflow/models
  OfficialIn papertf★ 77,694
• github.com/hebowei2000/deep-reinforcement-learning
  tf★ 7
• github.com/brandontrabucco/efficient-hrl
  tf★ 0
• github.com/AlexZhaoZt/Temporal_Leap_HRL
  tf★ 0
• github.com/sumkumar/hiro_impl
  tf★ 0
• github.com/tensorflow/models/tree/master/research/efficient-hrl
  tf★ 0

Abstract

We study the problem of representation learning in goal-conditioned hierarchical reinforcement learning. In such hierarchical structures, a higher-level controller solves tasks by iteratively communicating goals which a lower-level policy is trained to reach. Accordingly, the choice of representation -- the mapping of observation space to goal space -- is crucial. To study this problem, we develop a notion of sub-optimality of a representation, defined in terms of expected reward of the optimal hierarchical policy using this representation. We derive expressions which bound the sub-optimality and show how these expressions can be translated to representation learning objectives which may be optimized in practice. Results on a number of difficult continuous-control tasks show that our approach to representation learning yields qualitatively better representations as well as quantitatively better hierarchical policies, compared to existing methods (see videos at https://sites.google.com/view/representation-hrl).

Tasks

Reproductions