Pre-trained Language Models as Prior Knowledge for Playing Text-based Games

TL;DR

This paper introduces a framework combining pre-trained language models with deep reinforcement learning to improve agents' understanding and performance in text-based games, notably achieving higher scores on Zork1.

Contribution

The paper proposes integrating transformer-based language models with RL to enhance semantic understanding in text-based game agents, outperforming existing methods.

Findings

Outperforms all existing agents on Zork1 with a score of 44.7

Outperforms 4 out of 14 text-based games overall

Achieves comparable results to state-of-the-art on remaining games

Abstract

Recently, text world games have been proposed to enable artificial agents to understand and reason about real-world scenarios. These text-based games are challenging for artificial agents, as it requires an understanding of and interaction using natural language in a partially observable environment. Agents observe the environment via textual descriptions designed to be challenging enough for even human players. Past approaches have not paid enough attention to the language understanding capability of the proposed agents. Typically, these approaches train from scratch, an agent that learns both textual representations and the gameplay online during training using a temporal loss function. Given the sample-inefficiency of RL approaches, it is inefficient to learn rich enough textual representations to be able to understand and reason using the textual observation in such a complicated game environment setting. In this paper, we improve the semantic understanding of the agent by proposing a simple RL with LM framework where we use transformer-based language models with Deep RL models. We perform a detailed study of our framework to demonstrate how our model outperforms all existing agents on the popular game, Zork1, to achieve a score of 44.7, which is 1.6 higher than the state-of-the-art model. Overall, our proposed approach outperforms 4 games out of the 14 text-based games, while performing comparable to the state-of-the-art models on the remaining games.