Composing Task-Agnostic Policies with Deep Reinforcement Learning

Abstract

The composition of elementary behaviors to solve challenging transfer learning problems is one of the key elements in building intelligent machines. To date, there has been plenty of work on learning task-specific policies or skills but almost no focus on composing necessary, task-agnostic skills to find a solution to new problems. In this paper, we propose a novel deep reinforcement learning-based skill transfer and composition method that takes the agent's primitive policies to solve unseen tasks. We evaluate our method in difficult cases where training policy through standard reinforcement learning (RL) or even hierarchical RL is either not feasible or exhibits high sample complexity. We show that our method not only transfers skills to new problem settings but also solves the challenging environments requiring both task planning and motion control with high data efficiency.

arXiv commentICLR 2020

References in corpus (9)

• Continuous control with deep reinforcement learning
• Categorical Reparameterization with Gumbel-Softmax
• Trust Region Policy Optimization
• Addressing Function Approximation Error in Actor-Critic Methods
• Data-Efficient Hierarchical Reinforcement Learning
• Stochastic Neural Networks for Hierarchical Reinforcement Learning
• Intrinsically motivated reinforcement learning for human-robot interaction in the real-world
• Show, Attend and Interact: Perceivable Human-Robot Social Interaction through Neural Attention Q-Network
• Learning to Compose Skills

Cited by in corpus (5)

• Intelligent problem-solving as integrated hierarchical reinforcement learning
• Multi-Task Reinforcement Learning with Soft Modularization
• Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving
• Hierarchical principles of embodied reinforcement learning: A review
• Hierarchical Skills for Efficient Exploration