Happiness Pursuit: Personality Learning in a Society of Agents

TL;DR

This paper presents a method for learning distinct agent personalities using Deep Q-Networks trained on psychoanalytic rewards, analyzing happiness as a measure of alignment and overfitting in agent interactions within a game environment.

Contribution

It introduces a novel approach to personality learning in agents via DQN trained on psychoanalytic rewards, and studies happiness as an alignment metric in multi-agent interactions.

Findings

Higher happiness against hand-coded AI indicates overfitting.

Agents with higher happiness perform worse against other agents.

Happiness correlates with alignment and overfitting in agent interactions.

Abstract

Modeling personality is a challenging problem with applications spanning computer games, virtual assistants, online shopping and education. Many techniques have been tried, ranging from neural networks to computational cognitive architectures. However, most approaches rely on examples with hand-crafted features and scenarios. Here, we approach learning a personality by training agents using a Deep Q-Network (DQN) model on rewards based on psychoanalysis, against hand-coded AI in the game of Pong. As a result, we obtain 4 agents, each with its own personality. Then, we define happiness of an agent, which can be seen as a measure of alignment with agent's objective function, and study it when agents play both against hand-coded AI, and against each other. We find that the agents that achieve higher happiness during testing against hand-coded AI, have lower happiness when competing against each other. This suggests that higher happiness in testing is a sign of overfitting in learning to interact with hand-coded AI, and leads to worse performance against agents with different personalities.