Determining Game Quality Through UCT Tree Shape Analysis
نویسنده
چکیده
Upper Confidence Bounds for Trees (UCT) is an important new algorithm for tree searching that has shown itself to be very powerful. For example, in 2009, the Fuego software beat strong professional and amateur Go players by incorporating UCT and pattern recognition a feat which was previously thought to remain decades away. UCT has given rise to new fields of research such as Monte Carlo Tree Search. The hypothesis of this project is that there is a machine learnable relationship between the shape of UCT search trees for a variety of games their quality – how engaging the games are to humans. A statistically significant positive result was identified, representing a machine learning system that may be of great use when helping to classify automatically-generated board games in the future.
منابع مشابه
Towards Generalizing the Success of Monte-Carlo Tree Search beyond the Game of Go
Monte-Carlo Tree Search and specifically the variants of the UCT algorithm have been a break-through in AI of the board game Go. However, UCT has had limited applicability to other domains. We study the limitations of some of the existing variants of UCT in a small-scale Markov decision process (MDP), and propose new variants that can reduce those limitations. Our experiments show great improve...
متن کاملAn Automated Technique for Drafting Territories in the Board Game Risk
In the standard rules of the board game Risk, players take turns selecting or “drafting” the 42 territories on the board until all territories are owned. We present a technique for drafting territories in Risk that combines theMonte Carlo tree search algorithm UCT with an automated evaluation function. Created through supervised machine learning, this function scores outcomes of drafts in order...
متن کاملSmooth UCT Search in Computer Poker
Self-play Monte Carlo Tree Search (MCTS) has been successful in many perfect-information twoplayer games. Although these methods have been extended to imperfect-information games, so far they have not achieved the same level of practical success or theoretical convergence guarantees as competing methods. In this paper we introduce Smooth UCT, a variant of the established Upper Confidence Bounds...
متن کاملMonte-Carlo Planning for Pathfinding in Real-Time Strategy Games
In this work, we explore two Monte-Carlo planning approaches: Upper Confidence Tree (UCT) and Rapidlyexploring Random Tree (RRT). These Monte-Carlo planning approaches are applied in a real-time strategy game for solving the path finding problem. The planners are evaluated using a grid-based representation of our game world. The results show that the UCT planner solves the path planning problem...
متن کاملKnowledge Generation for Improving Simulations in UCT for General Game Playing
General Game Playing (GGP) aims at developing game playing agents that are able to play a variety of games and, in the absence of pre-programmed game specific knowledge, become proficient players. Most GGP players have used standard tree-search techniques enhanced by automatic heuristic learning. The UCT algorithm, a simulation-based tree search, is a new approach and has been used successfully...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010