• Publications
  • Influence
Monte-Carlo Tree Search: A New Framework for Game AI
TLDR
We put forward Monte-Carlo Tree Search as a novel, unified framework to game AI. Expand
  • 254
  • 28
  • PDF
Learning Tetris Using the Noisy Cross-Entropy Method
TLDR
The cross-entropy method is an efficient and general optimization algorithm. Expand
  • 203
  • 20
  • PDF
Model-based reinforcement learning with nearly tight exploration complexity bounds
TLDR
We show that Mormax, a modified version of the Rmax algorithm needs to make at most O(N log N) exploratory steps. Expand
  • 118
  • 11
  • PDF
Monte-Carlo Tree Search in Settlers of Catan
TLDR
We apply Monte-Carlo Tree Search (MCTS) to the multi-player, non-deterministic board game Settlers of Catan. Expand
  • 107
  • 9
  • PDF
Learning to Play Using Low-Complexity Rule-Based Policies: Illustrations through Ms. Pac-Man
TLDR
In this article we propose a method that can deal with certain combinatorial reinforcement learning tasks. Expand
  • 78
  • 7
  • PDF
Exploring compact reinforcement-learning representations with linear regression
TLDR
This paper presents a new algorithm for online linear regression whose efficiency guarantees satisfy the requirements of the KWIK (Knows What It Knows) framework. Expand
  • 90
  • 7
  • PDF
Reinforcement Learning with Echo State Networks
TLDR
We propose a novel method utilizing echo state networks (ESN) as function approximators in reinforcement learning. Expand
  • 50
  • 4
  • PDF
Cross-Entropy for Monte-Carlo Tree Search
TLDR
We investigate the use of the Cross-Entropy Method (CEM) (Rubinstein, 1999) for parameter tuning in general, i.e., for any game engine. Expand
  • 40
  • 3
  • PDF
Factored Value Iteration Converges
TLDR
In this paper we propose a novel algorithm, factored value iteration (FVI), for the approximate solution of factored Markov decision processes (fMDPs). Expand
  • 19
  • 3
  • PDF
MDPs: Learning in Varying Environments
TLDR
E-MDP-models are introduced and convergence theorems are proven using the generalized MDP framework of Szepesvari and Littman. Expand
  • 42
  • 2
  • PDF