Sökning: "spelkomplexitet"

Hittade 2 uppsatser innehållade ordet spelkomplexitet.

  1. 1. Exploring the effects of state-action space complexity on training time for AlphaZero agents

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Tobias Glimmerfors; [2022]
    Nyckelord :Deep learning; Reinforcement learning; AlphaZero; Monte-Carlo tree search; Environment complexity; Djupinlärning; Förstärkande inlärning; AlphaZero; Monte-Carlo tree search; spelkomplexitet;

    Sammanfattning : DeepMind’s development of AlphaGo took the world by storm in 2016 when it became the first computer program to defeat a world champion at the game of Go. Through further development, DeepMind showed that the underlying algorithm could be made more general, and applied to a large set of problems. LÄS MER

  3. 2. A comparison of two tree-search based algorithms for playing 3-dimensional Connect Four

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :David Avellan-Hultman; Emil Gunnberg Querat; [2021]
    Nyckelord :;

    Sammanfattning : This thesis aims to investigate general game-playing by conducting a comparison between the well-known methods Alpha-beta Pruning and Monte Carlo Tree Search in a new context, namely a three-dimensional version of the game Connect Four. The methods are compared by conducting a tournament with instances of both methods at varying levels of allowed search extent and measuring the performance as a function of the average thinking time taken per move. LÄS MER