THAIOct 11, 2023

Cheap Talking Algorithms

arXiv:2310.07867v62 citationsh-index: 1
Originality Incremental advance
AI Analysis

This demonstrates how simple learning algorithms can achieve equilibrium behavior in communication games, relevant for understanding algorithmic interactions in economics and AI.

The researchers simulated two independent reinforcement learning algorithms playing a strategic information transmission game, showing they converge to Nash equilibrium play with sender informativeness decreasing as bias increases, matching Pareto optimal or second-best equilibrium predictions at intermediate bias levels.

We simulate behaviour of two independent reinforcement learning algorithms playing the Crawford and Sobel (1982) game of strategic information transmission. We adopt memoryless algorithms to capture learning in a static game where a large population interacts anonymously. We show that sender and receiver converge to Nash equilibrium play. The level of informativeness of the sender's cheap talk decreases as the bias increases and, at intermediate level of the bias, it matches the level predicted by the Pareto optimal equilibrium or by the second best one. Conclusions are robust to alternative specifications of the learning hyperparameters and of the game.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes