Unsound Search with Policy and Value Networks in Legends of Code and Magic

• URL: http://arxiv.org/abs/2609.06816v1
• Date: Sun, 06 Sep 2026 20:18:52 GMT
• Title: Unsound Search with Policy and Value Networks in Legends of Code and Magic
• Abstract summary: Search is not a minor factor on the matchup between our agent and ByteRL.<n>We replicate a published best-response attack against ByteRL.<n>In LoCM unsound search gives us a stronger and more resilient agent.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Decision-time search in perfect and imperfect information games with enumerable belief states are effective methods for game AI. Collectible card games are imperfect information games with large belief states. Legends of Code and Magic is a collectible card game competition where the belief states are $2^{101}$. The Legends of Code and Magic (LoCM) champion, ByteRL, plays with no search. Other works claim sound enumeration-based search is unusable in the genre due to the number of belief states. We measured three previously defined properties that predict where theoretically unsound perfect information Monte Carlo's defects are cheap and found LoCM sits in the favorable region. Starting with imitation learning of the runner-up policy, NeteaseOPD, we created a policy and value feed-forward network. Our agent searches over worlds sampled from a prior over the opponent's deck built from the runner-up's drafts. Using our strictest configuration in the battle phase we beat ByteRL with a win percentage of 51.35% 95% CI [50.37, 52.33], over 10,000 pre-registered games using the LoCM official referee and time limit. Search is not a minor factor on the matchup between our agent and ByteRL. Without search this agent scores 26.8% and adding search adds +24.6 points. Unsound search in imperfect information games could be exploitable. We replicate a published best-response attack against ByteRL. We then apply the same attack protocol to two search configurations of our agent, and each one resists it better than ByteRL at every iteration. In LoCM unsound search gives us a stronger and more resilient agent.

Related papers

• MINDGAMES: A Live Arena for Evaluating Social and Strategic Reasoning in Multi-Agent LLMs [54.81359054218573]
  We introduce Mindgames, a multi-game arena and evaluation platform for large language models (LLMs)<n>Mindgames provides a unified interaction interface, TrueSkill-based rating, and full trajectory logging across four game environments.<n>We analyze 29,571 multi-agent games with turn-level observations, actions, and rewards, together with MG-Ref, a deterministic offline tournament protocol.
  arXiv  Detail & Related papers  (2026-05-28T07:33:47Z)
• Implicit Search via Discrete Diffusion: A Study on Chess [104.74301574891359]
  We propose DiffuSearch, a model that does textitimplicit search by looking into the future world via discrete diffusion modeling.<n>We instantiate DiffuSearch on a classical board game, Chess, where explicit search is known to be essential.<n>We show DiffuSearch outperforms both the searchless and explicit search-enhanced policies.
  arXiv  Detail & Related papers  (2025-02-27T06:25:15Z)
• Injecting Combinatorial Optimization into MCTS: Application to the Board Game boop [0.0]
  Combinatorial Optimization and Monte Carlo Tree Search can be combined efficiently. Our method beats 96% of the time the Monte Carlo Tree Search algorithm baseline. We opposed our AI method against human players on the Board Game Arena platform.
  arXiv  Detail & Related papers  (2024-06-13T02:55:08Z)
• Learning to Beat ByteRL: Exploitability of Collectible Card Game Agents [3.5036467860577307]
  We present preliminary analysis results of ByteRL, the state-of-the-art agent in Legends of Code and Magic and Hearthstone. Although ByteRL beat a top-10 Hearthstone player from China, we show that its play in Legends of Code and Magic is highly exploitable.
  arXiv  Detail & Related papers  (2024-04-25T15:48:40Z)
• Are AlphaZero-like Agents Robust to Adversarial Perturbations? [73.13944217915089]
  AlphaZero (AZ) has demonstrated that neural-network-based Go AIs can surpass human performance by a large margin. We ask whether adversarial states exist for Go AIs that may lead them to play surprisingly wrong actions. We develop the first adversarial attack on Go AIs that can efficiently search for adversarial states by strategically reducing the search space.
  arXiv  Detail & Related papers  (2022-11-07T18:43:25Z)
• Mastering the Game of No-Press Diplomacy via Human-Regularized Reinforcement Learning and Planning [95.78031053296513]
  No-press Diplomacy is a complex strategy game involving both cooperation and competition. We introduce a planning algorithm we call DiL-piKL that regularizes a reward-maximizing policy toward a human imitation-learned policy. We show that DiL-piKL can be extended into a self-play reinforcement learning algorithm we call RL-DiL-piKL.
  arXiv  Detail & Related papers  (2022-10-11T14:47:35Z)
• Developing a Successful Bomberman Agent [0.6606016007748989]
  We compare the behavior of three search algorithms: Monte Carlo Tree Search, Rolling Horizon Evolution, and Beam Search. We present various enhancements leading to improve the agents' strength that concern search, opponent prediction, game state evaluation, and game engine encoding.
  arXiv  Detail & Related papers  (2022-03-17T20:47:52Z)
• Learning to Stop: Dynamic Simulation Monte-Carlo Tree Search [66.34387649910046]
  Monte Carlo tree search (MCTS) has achieved state-of-the-art results in many domains such as Go and Atari games. We propose to achieve this goal by predicting the uncertainty of the current searching status and use the result to decide whether we should stop searching.
  arXiv  Detail & Related papers  (2020-12-14T19:49:25Z)
• Single-Agent Optimization Through Policy Iteration Using Monte-Carlo Tree Search [8.22379888383833]
  Combination of Monte-Carlo Tree Search (MCTS) and deep reinforcement learning is state-of-the-art in two-player perfect-information games. We describe a search algorithm that uses a variant of MCTS which we enhanced by 1) a novel action value normalization mechanism for games with potentially unbounded rewards, 2) defining a virtual loss function that enables effective search parallelization, and 3) a policy network, trained by generations of self-play, to guide the search.
  arXiv  Detail & Related papers  (2020-05-22T18:02:36Z)
• Efficient exploration of zero-sum stochastic games [83.28949556413717]
  We investigate the increasingly important and common game-solving setting where we do not have an explicit description of the game but only oracle access to it through gameplay. During a limited-duration learning phase, the algorithm can control the actions of both players in order to try to learn the game and how to play it well. Our motivation is to quickly learn strategies that have low exploitability in situations where evaluating the payoffs of a queried strategy profile is costly.
  arXiv  Detail & Related papers  (2020-02-24T20:30:38Z)
• Deep RL Agent for a Real-Time Action Strategy Game [0.3867363075280543]
  We introduce a reinforcement learning environment based on Heroic - Magic Duel, a 1 v 1 action strategy game. Our main contribution is a deep reinforcement learning agent playing the game at a competitive level. Our best self-play agent, obtains around $65%$ win rate against the existing AI and over $50%$ win rate against a top human player.
  arXiv  Detail & Related papers  (2020-02-15T01:09:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.