Team learning from human demonstration with coordination confidence
- 5 November 2019
- journal article
- research article
- Published by Maximum Academic Press in The Knowledge Engineering Review
- Vol. 34
Abstract
Among an array of techniques proposed to speed-up reinforcement learning (RL), learning from human demonstration has a proven record of success. A related technique, called Human-Agent Transfer, and its confidence-based derivatives have been successfully applied to single-agent RL. This article investigates their application to collaborative multi-agent RL problems. We show that a first-cut extension may leave room for improvement in some domains, and propose a new algorithm called coordination confidence (CC). CC analyzes the difference in perspectives between a human demonstrator (global view) and the learning agents (local view) and informs the agents’ action choices when the difference is critical and simply following the human demonstration can lead to miscoordination. We conduct experiments in three domains to investigate the performance of CC in comparison with relevant baselines.Keywords
This publication has 8 references indexed in Scilit:
- Interactive Reinforcement Learning with Dynamic Reuse of Prior Knowledge from Human and Agent Demonstrations2019
- Improving Reinforcement Learning with Confidence-Based Demonstrations2017
- Multi-agent reinforcement learning as a rehearsal for decentralized planningNeurocomputing, 2016
- Mastering the game of Go with deep neural networks and tree searchNature, 2016
- Human-level control through deep reinforcement learningNature, 2015
- Probabilistic Policy Reuse for inter-task transfer learningRobotics and Autonomous Systems, 2010
- A survey of robot learning from demonstrationRobotics and Autonomous Systems, 2009
- Reinforcement Learning: An IntroductionIEEE Transactions on Neural Networks, 1998