Hierarchical Reinforcement Learning By Discovering Intrinsic Options

Abstract

We propose a hierarchical reinforcement learning method, HIDIO, that can learn task-agnostic options in a self-supervised manner while jointly learning to utilize them to solve sparse-reward tasks. Unlike current hierarchical RL approaches that tend to formulate goal-reaching low-level tasks or pre-define ad hoc lower-level policies, HIDIO encourages lower-level option learning that is independent of the task at hand, requiring few assumptions or little knowledge about the task structure. These options are learned through an intrinsic entropy minimization objective conditioned on the option sub-trajectories. The learned options are diverse and task-agnostic. In experiments on sparse-reward robotic manipulation and navigation tasks, HIDIO achieves higher success rates with greater sample efficiency than regular RL baselines and two state-of-the-art hierarchical RL methods.

arXiv commentICLR 2021. 19 pages, 9 figures. Code at https://www.github.com/jesbu1/hidio

References in corpus (16)

• Continuous control with deep reinforcement learning
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Learning by Playing - Solving Sparse Reward Tasks from Scratch
• FeUdal Networks for Hierarchical Reinforcement Learning
• Data-Efficient Hierarchical Reinforcement Learning
• Variational Intrinsic Control
• Revisiting Fundamentals of Experience Replay
• Efficient Parallel Methods for Deep Reinforcement Learning
• Learning Multi-Level Hierarchies with Hindsight
• Dynamics-Aware Unsupervised Discovery of Skills
• Hierarchical visuomotor control of humanoids
• Unsupervised Control Through Non-Parametric Discriminative Rewards
• Learning Abstract Options
• Directed-Info GAIL: Learning Hierarchical Policies from Unsegmented Demonstrations using Directed Information
• Hierarchical Reinforcement Learning for Zero-shot Generalization with Subtask Dependencies
• Unsupervised Curricula for Visual Meta-Reinforcement Learning

Cited by in corpus (6)

• An information-theoretic perspective on intrinsic motivation in reinforcement learning: a survey
• TAAC: Temporally Abstract Actor-Critic for Continuous Control
• Value Function Spaces: Skill-Centric State Abstractions for Long-Horizon Reasoning
• Hierarchical Skills for Efficient Exploration
• HAC Explore: Accelerating Exploration with Hierarchical Reinforcement Learning
• Direct then Diffuse: Incremental Unsupervised Skill Discovery for State Covering and Goal Reaching