Designing an Automatic Agent for Repeated Language based Persuasion Games

TL;DR

This paper introduces an automatic agent for repeated language-based persuasion games, leveraging deep learning and Monte Carlo Tree Search to optimize persuasive reviews in natural language, outperforming baselines.

Contribution

It presents a novel MCTS-based framework with deep learning models for natural language persuasion in repeated economic games, addressing a gap in language-rich communication modeling.

Findings

The expert outperforms strong baselines.

It adapts effectively to different decision makers.

Selected reviews are well tailored to the deals.

Abstract

Persuasion games are fundamental in economics and AI research and serve as the basis for important applications. However, work on this setup assumes communication with stylized messages that do not consist of rich human language. In this paper we consider a repeated sender (expert) -- receiver (decision maker) game, where the sender is fully informed about the state of the world and aims to persuade the receiver to accept a deal by sending one of several possible natural language reviews. We design an automatic expert that plays this repeated game, aiming to achieve the maximal payoff. Our expert is implemented within the Monte Carlo Tree Search (MCTS) algorithm, with deep learning models that exploit behavioral and linguistic signals in order to predict the next action of the decision maker, and the future payoff of the expert given the state of the game and a candidate review. We demonstrate the superiority of our expert over strong baselines, its adaptability to different decision makers, and that its selected reviews are nicely adapted to the proposed deal.