Imitating Interactive Intelligence
Abstract
A common vision from science fiction is that robots will one day inhabit our physical spaces, sense the world as we do, assist our physical labours, and communicate with us through natural language. Here we study how to design artificial agents that can interact naturally with humans using the simplification of a virtual environment. This setting nevertheless integrates a number of the central challenges of artificial intelligence (AI) research: complex visual perception and goal-directed physical control, grounded language comprehension and production, and multi-agent social interaction. To build agents that can robustly interact with humans, we would ideally train them while they interact with humans. However, this is presently impractical. Therefore, we approximate the role of the human with another learned agent, and use ideas from inverse reinforcement learning to reduce the disparities between human-human and agent-agent interactive behaviour. Rigorously evaluating our agents poses a great challenge, so we develop a variety of behavioural tests, including evaluation by humans who watch videos of agents or interact directly with them. These evaluations convincingly demonstrate that interactive training and auxiliary losses improve agent behaviour beyond what is achieved by supervised learning of actions alone. Further, we demonstrate that agent capabilities generalise beyond literal experiences in the dataset. Finally, we train evaluation models whose ratings of agents agree well with human judgement, thus permitting the evaluation of new agent models without additional effort. Taken together, our results in this virtual environment provide evidence that large-scale human behavioural imitation is a promising tool to create intelligent, interactive agents, and the challenge of reliably evaluating such agents is possible to surmount.
Cite
@article{arxiv.2012.05672,
title = {Imitating Interactive Intelligence},
author = {Josh Abramson and Arun Ahuja and Iain Barr and Arthur Brussee and Federico Carnevale and Mary Cassin and Rachita Chhaparia and Stephen Clark and Bogdan Damoc and Andrew Dudzik and Petko Georgiev and Aurelia Guy and Tim Harley and Felix Hill and Alden Hung and Zachary Kenton and Jessica Landon and Timothy Lillicrap and Kory Mathewson and Soňa Mokrá and Alistair Muldal and Adam Santoro and Nikolay Savinov and Vikrant Varma and Greg Wayne and Duncan Williams and Nathaniel Wong and Chen Yan and Rui Zhu},
journal= {arXiv preprint arXiv:2012.05672},
year = {2021}
}