The Intentional Unintentional Agent: Learning to Solve Many Continuous Control Tasks Simultaneously

Abstract

This paper introduces the Intentional Unintentional (IU) agent. This agent endows the deep deterministic policy gradients (DDPG) agent for continuous control with the ability to solve several tasks simultaneously. Learning to solve many tasks simultaneously has been a long-standing, core goal of artificial intelligence, inspired by infant development and motivated by the desire to build flexible robot manipulators capable of many diverse behaviours. We show that the IU agent not only learns to solve many tasks simultaneously but it also learns faster than agents that target a single task at-a-time. In some cases, where the single task DDPG method completely fails, the IU agent successfully solves the task. To demonstrate this, we build a playroom environment using the MuJoCo physics engine, and introduce a grounded formal language to automatically generate tasks.

References in corpus (6)

• Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
• Learning to Execute
• Hybrid Reward Architecture for Reinforcement Learning
• Grounded Language Learning in a Simulated 3D World
• Recurrent Reinforcement Learning: A Hybrid Approach
• Programmable Agents

Cited by in corpus (9)

• Learning by Playing - Solving Sparse Reward Tasks from Scratch
• Grasp2Vec: Learning Object Representations from Self-Supervised Grasping
• Computational Theories of Curiosity-Driven Learning
• Learning Latent Plans from Play
• Learning latent state representation for speeding up exploration
• Generalized Hindsight for Reinforcement Learning
• Data-efficient Hindsight Off-policy Option Learning
• Hierarchical principles of embodied reinforcement learning: A review
• Scaling All-Goals Updates in Reinforcement Learning Using Convolutional Neural Networks