ThreePlayerIPD

Description

ThreePlayerIPD is an environment for evaluating agents on multi-player game theory and social dilemmas in the Iterated Prisoner's Dilemma. This environment wraps the ThreePlayerIPD implementation from TextArena, a framework for text-based game environments.

Capabilities

• Strategic decision-making under uncertainty
• Communication and persuasion in game-theoretic contexts
• Trust building and defection detection
• Multi-party cooperation and competition dynamics

Compute Requirements

ThreePlayerIPD does not require a sandbox. It has minimal compute requirements.

License

MIT.

Tasks

There are two splits: train (150 tasks) and test (150 tasks). Each split contains 50 tasks across each of 3 variants:

• ThreePlayerIPD-v0
• ThreePlayerIPD-v0-train
• ThreePlayerIPD-v0-raw

Each task is seeded for reproducibility.

Reward Structure

This is a sparse reward environment. Rewards are mapped from TextArena's native range of {-1, 0, 1} to {0.0, 0.5, 1.0} via (raw + 1) / 2.

We do not use LLM graders for this environment; reward is determined programmatically.

Data

Game state is generated procedurally by the TextArena engine using seeded randomness. No external data files are required.

Tools

Agents are given a single tool:

• send_message(message): Send a chat message or submit your decision. For decisions use format: '[1 cooperate] [2 defect]'

Time Horizon

ThreePlayerIPD is a multi-turn environment.

Environment Difficulty

Medium-High. Agents must balance short-term incentives with long-term cooperation, communicate effectively, and adapt strategies based on opponents' behavior across 5 rounds of chat and decision phases.

Other Environment Requirements

This environment requires an OpenAI API key (passed via secrets) to power the LLM opponents.

Safety

Agents in ThreePlayerIPD interact only with a game-theoretic simulation and have no access to external systems, the internet, or sensitive data. The environment does not present safety risks.

Citations

@software{textarena2024,
  author    = {Guertler, Leon and Banting, Wilfried and Pignatelli, Eduardo},
  title     = {TextArena},
  year      = {2024},
  publisher = {GitHub},
  url       = {https://github.com/LeonGuertler/TextArena}
}