Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Valet: A Standardized Testbed of Traditional Imperfect-Information Card Games

Mark Goadrich, Achille Morenville, Éric Piette

TL;DR

This work introduces Valet, a diverse and comprehensive testbed of 21 traditional imperfect-information card games, and empirically characterize each game's branching factor and duration using random simulations, reporting baseline score distributions for a Monte Carlo Tree Search player against random opponents to demonstrate the suitability of Valet as a benchmarking suite.

Abstract

AI algorithms for imperfect-information games are typically compared using performance metrics on individual games, making it difficult to assess robustness across game choices. Card games are a natural domain for imperfect information due to hidden hands and stochastic draws. To facilitate comparative research on imperfect-information game-playing algorithms and game systems, we introduce Valet, a diverse and comprehensive testbed of 21 traditional imperfect-information card games. These games span multiple genres, cultures, player counts, deck structures, mechanics, winning conditions, and methods of hiding and revealing information. To standardize implementations across systems, we encode the rules of each game in RECYCLE, a card game description language. We empirically characterize each game's branching factor and duration using random simulations, reporting baseline score distributions for a Monte Carlo Tree Search player against random opponents to demonstrate the suitability of Valet as a benchmarking suite.

Valet: A Standardized Testbed of Traditional Imperfect-Information Card Games

TL;DR

This work introduces Valet, a diverse and comprehensive testbed of 21 traditional imperfect-information card games, and empirically characterize each game's branching factor and duration using random simulations, reporting baseline score distributions for a Monte Carlo Tree Search player against random opponents to demonstrate the suitability of Valet as a benchmarking suite.

Abstract

AI algorithms for imperfect-information games are typically compared using performance metrics on individual games, making it difficult to assess robustness across game choices. Card games are a natural domain for imperfect information due to hidden hands and stochastic draws. To facilitate comparative research on imperfect-information game-playing algorithms and game systems, we introduce Valet, a diverse and comprehensive testbed of 21 traditional imperfect-information card games. These games span multiple genres, cultures, player counts, deck structures, mechanics, winning conditions, and methods of hiding and revealing information. To standardize implementations across systems, we encode the rules of each game in RECYCLE, a card game description language. We empirically characterize each game's branching factor and duration using random simulations, reporting baseline score distributions for a Monte Carlo Tree Search player against random opponents to demonstrate the suitability of Valet as a benchmarking suite.
Paper Structure (13 sections, 4 figures, 2 tables)

This paper contains 13 sections, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Testbed Curation
  3. Genre
  4. Deck Composition
  5. Variants and RECYCLE Encoding
  6. Evaluation
  7. Information Flow
  8. Branching Factor
  9. Game Length
  10. Score Distribution
  11. Discussion and Conclusion

Figures (4)

  • Figure 1: Information flow diagrams for six games in Valet.
  • Figure 2: The branching factor at each decision point of the game, summarized from 100 random rollouts. Each player is denoted with a different color. Since both axis are integers, an x-y jitter filter is used to help visualize the distributions.
  • Figure 3: The distribution of game lengths over 100 random rollouts for each game. The lengths of the game are on a logarithmic scale.
  • Figure 4: A violin plot of the distribution of scores for the first player in each game, comparing MCTS with random.