An Extensible Interactive Interface for Agent Design

Abstract

In artificial intelligence, we often specify tasks through a reward function. While this works well in some settings, many tasks are hard to specify this way. In deep reinforcement learning, for example, directly specifying a reward as a function of a high-dimensional observation is challenging. Instead, we present an interface for specifying tasks interactively using demonstrations. Our approach defines a set of increasingly complex policies. The interface allows the user to switch between these policies at fixed intervals to generate demonstrations of novel, more complex, tasks. We train new policies based on these demonstrations and repeat the process. We present a case study of our approach in the Lunar Lander domain, and show that this simple approach can quickly learn a successful landing policy and outperforms an existing comparison-based deep RL method.

arXiv commentPresented at 2019 ICML Workshop on Human in the Loop Learning (HILL 2019), Long Beach, USA

References in corpus (5)

• Visual Reinforcement Learning with Imagined Goals
• Grounded Language Learning in a Simulated 3D World
• Learning to Understand Goal Specifications by Modelling Reward
• Variational Inverse Control with Events: A General Framework for Data-Driven Reward Definition
• Guiding Policies with Language via Meta-Learning