boardgame-research

This is a list of boardgame research. They are primarily related to "solving/playing/learning" games (by various different approaches), or occasionaly about designing or meta-aspects of the game. This doesn't cover all aspects of each game (notably missing social-science stuff), but should be of interest to anyone interested in boardgames and their optimal play. While there is a ton of easily accessible research on games like Chess and Go, finding prior work on more contemporary games can be a bit hard. This list focuses on the latter. If you are interested in well-researched games like Chess, Go, Hex, take a look at the Chess programming wiki instead. The list also covers some computer-games that fall under similar themes.

Exported versions (work-in-progress) are available in the following formats:

• Zotero RDF
• BibTeX
• HTML Bookmarks

Subscribe to the repo to get the latest updates for now.

If you aren't able to access any paper on this list, please try using Sci-Hub or reach out to me.

Table of Contents generated with DocToc

• Settlers of Catan
• Modern Art: The card game
• Diplomacy
• Risk
• Kingdomino
• Patchwork
• Nmbr9
• Hanabi
• Azul
• Monopoly
• Magic: the Gathering
• Terra Mystica
• Mafia
• The Resistance: Avalon
• Ticket to Ride
• Lost Cities
• Uno
• Dominion
• Quixo
• Race for the Galaxy
• Santorini
• Set
• Pentago
• Blokus
• Pandemic
• Scotland Yard
• Monopoly Deal
• Yahtzee
• Mobile Games
  • 2048
• Game Design
• Frameworks/Toolkits

Settlers of Catan

• The effectiveness of persuasion in The Settlers of Catan
• Avoiding Revenge Using Optimal Opponent Ranking Strategy in the Board Game Catan
• Game strategies for The Settlers of Catan
• Monte-Carlo Tree Search in Settlers of Catan
• Settlers of Catan bot trained using reinforcement learning (MATLAB).
• Trading in a multiplayer board game: Towards an analysis of non-cooperative dialogue
• POMCP with Human Preferencesin Settlers of Catan
• Reinforcement Learning of Strategies for Settlers of Catan
• Deep Reinforcement Learning in Strategic Board GameEnvironments [pdf]
• Monte Carlo Tree Search in a Modern Board Game Framework
• The impact of loaded dice in Catan

Modern Art: The card game

• A constraint programming based solver for Modern Art

Diplomacy

• Learning to Play No-Press Diplomacy with Best Response Policy Iteration
• No Press Diplomacy: Modeling Multi-Agent Gameplay
• Agent Madoff: A Heuristic-Based Negotiation Agent For The Diplomacy Strategy Game

Risk

• Mini-Risk: Strategies for a Simplified Board Game
• A Multi-Agent System for playing the board game Risk
• Learning the risk board game with classifier systems
• Markov Chains and the RISK Board Game
• Markov Chains for the RISK Board Game Revisited
• RISK Board Game ‐ Battle Outcome Analysis
• Planning an endgame move set for the game RISK
• RISKy Business: An In-Depth Look at the Game RISK
• An Intelligent Artificial Player for the Game of Risk

Kingdomino

• Monte Carlo Methods for the Game Kingdomino [arXiv]
• NP-completeness of the game Kingdomino

Patchwork

• State Representation and Polyomino Placement for the Game Patchwork

Nmbr9

• Nmbr9 as a Constraint Programming Challenge

Hanabi

• Improving Policies via Search in Cooperative Partially Observable Games (FB) [code] - Current best result.
• Re-determinizing MCTS in Hanabi
• Hanabi is NP-hard, Even for Cheaters who Look at Their Cards
• Evolving Agents for the Hanabi 2018 CIG Competition
• Aspects of the Cooperative Card Game Hanabi
• How to Make the Perfect Fireworks Display: Two Strategies for Hanabi
• Playing Hanabi Near-Optimally
• Evaluating and modelling Hanabi-playing agents
• An intentional AI for hanabi
• The Hanabi challenge: A new frontier for AI research [arXiv]] (DeepMind)
• Solving Hanabi: Estimating Hands by Opponent's Actions in Cooperative Game with Incomplete Information
• A Browser-based Interface for the Exploration and Evaluation of Hanabi AIs
• I see what you see: Integrating eye tracking into Hanabi playing agents
• The 2018 Hanabi competition
• Diverse Agents for Ad-Hoc Cooperation in Hanabi [arXiv]
• State of the art Hanabi bots + simulation framework in rust
• A strategy simulator for the well-known cooperative card game Hanabi
• A framework for writing bots that play Hanabi
• Evaluating the Rainbow DQN Agent in Hanabi with Unseen Partners

Azul

• A summary of a dissertation on Azul (unpublished)

Monopoly

• Negotiation strategy of agents in the MONOPOLY game
• Generating interesting Monopoly boards from open data
• Estimating the probability that the game of Monopoly never ends
• Learning to play Monopoly:A Reinforcement Learning approach
• Monopoly as a Markov Process
• Learning to Play Monopoly withMonte Carlo Tree Search
• Monopoly Using Reinforcement Learning
• A Markovian Exploration of Monopoly
• What's the best Monopoly strategy

Magic: the Gathering

• Magic: the Gathering is as Hard as Arithmetic
• Magic: The Gathering is Turing Complete
• Neural Networks Models for Analyzing Magic: the Gathering Cards [arXiv]
• The Complexity of Deciding Legality of a Single Step of Magic: the Gathering
• Ensemble Determinization in Monte Carlo Tree Search for the Imperfect Information Card Game Magic: The Gathering
• Magic: The Gathering in Common Lisp [source]
• Deck Costruction Strategies for Magic: the Gathering
• Deckbuilding in Magic: The Gathering Using a Genetic Algorithm
• Mathematical programming and Magic: The Gathering®
• Optimal Card-Collecting Strategies for Magic: The Gathering
• Monte Carlo search applied to card selection in Magic: The Gathering
• Magic: The Gathering Deck Performance Prediction

Terra Mystica

• Using Tabu Search Algorithm for Map Generation in the Terra Mystica Tabletop Game

Mafia

• A mathematical model of the Mafia game
• Automatic Long-Term Deception Detection in Group Interaction Videos
• Human-Side Strategies in the Werewolf Game Against the Stealth Werewolf Strategy

The Resistance: Avalon

• Finding Friend and Foe in Multi-Agent Games

Ticket to Ride

• Evolving maps and decks for ticket to ride
• Materials for Ticket to Ride Seattle and a framework for making more game boards
• Applications of Graph Theory andProbability in the Board GameTicket toRide
• The Difficulty of Learning Ticket to Ride

Lost Cities

• Applying Neural Networks and Genetic Programming to the Game Lost Cities

Uno

• The complexity of UNO
• UNO Is Hard, Even for a Single Player

Dominion

There is a simulator and the code behind the Dominion server running councilroom.com is available. councilroom has the best and worst openings, optimal card ratios, Card winning stats and lots of other empirical research.

• Clustering Player Strategies from Variable-Length Game Logs in Dominion

Quixo

• QUIXO is EXPTIME-complete

Race for the Galaxy

• SCOUT: A Case-Based Reasoning Agent for Playing Race for the Galaxy

Santorini

• A Mathematical Analysis of the Game of Santorini

Set

Set has a long history of mathematical research, so this list isn't exhaustive.

• Game, Set, Math
• The Joy of SET

Pentago

• On Solving Pentago

Blokus

• Blokus Game Solver

Pandemic

• NP-Completeness of Pandemic

Scotland Yard

• The complexity of Scotland Yard

Monopoly Deal

• Implementation of AI Player on "Monopoly Deal"

Yahtzee

• Optimal Solitaire Yahtzee Strategies
• Nearly Optimal Computer Play in Multi-player Yahtzee
• Computer Strategies for Solitaire Yahtzee
• An optimal strategy for Yahtzee
• Yahtzee: a Large Stochastic Environment for RL Benchmarks
• Modeling expert problem solving in a game of chance: a Yahtzee case study
• Probabilites In Yahtzee
• Optimal Yahtzee performance in multi-player games
• Defensive Yahtzee
• Using Deep Q-Learning to Compare Strategy Ladders of Yahtzee
• How to Maximize Your Score in Solitaire Yahtzee

Mobile Games

• Trainyard is NP-Hard
• Threes!, Fives, 1024!, and 2048 are Hard

2048

• Making Change in 2048
• Analysis of the Game "2048" and its Generalization in Higher Dimensions
• Temporal difference learning of N-tuple networks for the game 2048
• Multi-Stage Temporal Difference Learning for 2048-like Games
• On the Complexity of Slide-and-Merge Games
• 2048 is (PSPACE) Hard, but Sometimes Eas
• Systematic Selection of N-Tuple Networks for 2048
• Systematic selection of N-tuple networks with consideration of interinfluence for game 2048
• 2048 Without New Tiles Is Still Hard
• An investigation into 2048 AI strategies

Game Design

• MDA: A Formal Approach to Game Design and Game Research
• Exploring Anonymity in Cooperative Board Games

Frameworks/Toolkits

• RLCard: A Toolkit for Reinforcement Learning in Card Games
• GTSA: Game Tree Search Algorithms