Parallel Monte Carlo Tree Search on GPU

  title={Parallel Monte Carlo Tree Search on GPU},
  author={K. Rocki and R. Suda},
Monte Carlo Tree Search (MCTS) is a method for making optimal decisions in artificial intelligence (AI) problems, typically move planning in combinatorial games. It combines the generality of random simulation with the precision of tree search. It can theoretically be applied to any domain that can be described in terms of state, action pairs and simulation used to forecast outcomes such as decision support, control, delayed reward problems or complex optimization. The motivation behind this… Expand
Large-Scale Parallel Monte Carlo Tree Search on GPU
  • K. Rocki, R. Suda
  • Computer Science
  • 2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum
  • 2011
An efficient parallel GPU MCTS implementation based on the introduced 'block-parallelism' scheme which combines GPU SIMD thread groups and performs independent searches without any need of intra-GPU or inter-GPU communication is presented. Expand
Application and Design of GPU Parallel RRT for Racing Car Simulation
Graphical Processing Units (GPUs) have evolved at a large pace, maintaining a processing power orders of magnitude higher than Central Processing Units (CPUs). As a result, the interest of using theExpand
Iterative Parallel Sampling RRT for Racing Car Simulation
A new variant of the RRT algorithm called Iterative Parallel Sampling RRT is proposed which explores the use of parallel computation in GPU to generate faster solutions and has the potential for reducing search times. Expand
Leveraging Parallel Architectures in AI Search Algorithms for Games
This paper proposes to take advantage of different parallel architectures to help solve the problems of parallelizing UCT search on GPUs, the development of a hierarchical search framework for Real-Time Strategy (RTS) games and the building placement problem in RTS games. Expand
Parallel UCT search on GPUs
Empirical results show that the proposed Multiblock Parallel algorithm outperforms other approaches and can take advantage of the GPU hardware without the added complexity of searching multiple trees. Expand


Parallel Monte-Carlo Tree Search
Three parallelization methods for MCTS are discussed: leaf parallelization, root Parallelization, and tree parallelization. Expand
Monte-Carlo tree search in production management problems
It is shown that Monte-Carlo Tree Search leads to a solution in a shorter period of time than this algorithm, with improved solutions for large problems. Expand
Efficient Selectivity and Backup Operators in Monte-Carlo Tree Search
  • Rémi Coulom
  • Mathematics, Computer Science
  • Computers and Games
  • 2006
A new framework to combine tree search with Monte-Carlo evaluation, that does not separate between a min-max phase and a Monte- carlo phase is presented, that provides finegrained control of the tree growth, at the level of individual simulations, and allows efficient selectivity. Expand
Massively Parallel Monte Carlo Tree Search
Monte Carlo Tree Search is a method of finding near-optimal solutions for large state-space problems. Currently, is it very important to develop algorithms being able to take advantage of greatExpand
Feature Selection as a One-Player Game
This paper formalizes Feature Selection as a Reinforcement Learning problem, leading to a provably optimal though intractable selection policy, and presents an approximation thereof, based on a one-player game approach and relying on the Monte-Carlo tree search UCT (Upper Confidence Tree) proposed by Kocsis and Szepesvari (2006). Expand
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. Expand
Hacker's Delight
The term "hacker" in the title is meant in the originalsense of an aficionado of computers—someone who enjoys making computers do new things, or do old things in a new and clever way. Expand
Jaap van den Herik: Parallel Monte-Carlo Tree Search
  • Computers and Games: 6th International Conference
  • 2008
High-dimensional planning with Monte-Carlo Tree Search
  • High-dimensional planning with Monte-Carlo Tree Search
  • 2008
High-dimensional planning with Monte-Carlo Tree Search
  • 2008