Mastering Strategy Card Game (Legends of Code and Magic) via End-to-End Policy and Optimistic Smooth Fictitious Play

Wei Xi, Yongxin Zhang, Changnan Xiao, Xuefeng Huang, Shihong Deng +3 more
3/7/2023
cs.LGcs.AIcs.GT

No official code found. Best implementation: ronaldosvieira/gym-locm (39 stars, MIT, last commit 2026-04-28).

Abstract

Deep Reinforcement Learning combined with Fictitious Play shows impressive results on many benchmark games, most of which are, however, single-stage. In contrast, real-world decision making problems may consist of multiple stages, where the observation spaces and the action spaces can be completely different across stages. We study a two-stage strategy card game Legends of Code and Magic and propose an end-to-end policy to address the difficulties that arise in multi-stage game. We also propose an optimistic smooth fictitious play algorithm to find the Nash Equilibrium for the two-player game. Our approach wins double championships of COG2022 competition. Extensive studies verify and show the advancement of our approach.

View on arXivView PDF

Code Implementations(1)

ronaldosvieira/gym-locmCommunity100%
MIT

Gymnasium environments for Legends of Code and Magic, a collectible card game designed for AI research

395PythonJun 4, 20195 months agoMIT
collectible-card-gamesgym-environmentlegends-of-code-and-magicreinforcement-learning

Similar Papers

AlphaDou: High-Performance End-to-End Doudizhu AI Integrating Bidding

85% match1 repo532024

Mastering Strategy Card Game (Hearthstone) with Improved Techniques

84% match2023

DouRN: Improving DouZero by Residual Neural Networks

83% match2024

Reinforcing Competitive Multi-Agents for Playing 'So Long Sucker'

83% match2024

Computing Optimal Equilibria and Mechanisms via Learning in Zero-Sum Extensive-Form Games

82% match2026

SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning

81% match1 repo2042025

Cite this paper

@article{xi2023mastering,
  title  = {Mastering Strategy Card Game (Legends of Code and Magic) via End-to-End Policy and Optimistic Smooth Fictitious Play},
  author = {Wei Xi and Yongxin Zhang and Changnan Xiao and Xuefeng Huang and Shihong Deng and Haowei Liang and Jie Chen and Peng Sun},
  year   = {2023},
  eprint = {2303.04096},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2303.04096v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2303.04096.svg)](https://codeofpaper.com/paper/2303.04096)

Discussion