Optimizing UCT for Settlers of Catan

  title={Optimizing UCT for Settlers of Catan},
  author={Gabriel de Arruda Rubin de Lima and Bruno Fortes Paz and Felipe Meneguzzi},
Settlers of Catan is one of the main representatives of modern strategic board games and there are few autonomous agents available to play it due to its challenging features such as stochasticity, imperfect information, and 4-player structure. In this paper, we extend previous work on UCT search to develop an automated player for Settlers of Catan. Specifically, we develop a move pruning heuristic for this game and introduce the ability to trade with the other players using the UCT algorithm… 

Figures and Tables from this paper



Monte-Carlo Tree Search in Settlers of Catan

It is shown that MCTS can be adapted successfully to multi-agent environments, and the results show that the agent has a considerable playing strength when compared to game implementation with existing heuristics.

Game strategies for The Settlers of Catan

This framework provides the means to change different components of an autonomous agent's strategy, and to test them in suitably controlled ways via performance metrics in game simulations and via comparisons of the agent's behaviours with those exhibited in a corpus of humans playing the game.

Towards Human-Competitive Game Playing for Complex Board Games with Genetic Programming

This work shows that Genetic Programming GP can evolve competitive artificial intelligence AI players against human-designed AI or against Monte Carlo Tree Search, a standard in automatic game playing.

Lower Bounding Klondike Solitaire with Monte-Carlo Planning

This paper studies Klondike using several sampling-based planning approaches including UCT, hindsight optimization, and sparse sampling, and establishes empirical lower bounds on their performance and provides a theoretical bound on the sample complexity of a method that naturally combines sparse sampling and UCT.

Mastering the game of Go with deep neural networks and tree search

Using this search algorithm, the program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0.5, the first time that a computer program has defeated a human professional player in the full-sized game of Go.

Finite-time Analysis of the Multiarmed Bandit Problem

This work shows that the optimal logarithmic regret is also achievable uniformly over time, with simple and efficient policies, and for all reward distributions with bounded support.

Real-time decision making for adversarial environments using a plan-based heuristic

In designing this agent the goal was to create a control technique that allowed the agent to decide a course of action given its current goals and situation, and by using a computer game environment the agent was forced to address some of the real world constraints faced by computer game producers.

A Survey of Monte Carlo Tree Search Methods

A survey of the literature to date of Monte Carlo tree search, intended to provide a snapshot of the state of the art after the first five years of MCTS research, outlines the core algorithm's derivation, impart some structure on the many variations and enhancements that have been proposed, and summarizes the results from the key game and nongame domains.

Exploration exploitation in Go: UCT for Monte-Carlo Go

A Monte-Carlo program, MoGo, is developed, which is the first computer Go program using UCT, among which efficient memory management, parametrization, ordering of non-visited nodes and parallelization are explained.

Bandit Based Monte-Carlo Planning

A new algorithm is introduced, UCT, that applies bandit ideas to guide Monte-Carlo planning and is shown to be consistent and finite sample bounds are derived on the estimation error due to sampling.