Advances in Computer Games

1. Front Matter

   Pages I-XX

   PDF

2. Adaptive Playouts in Monte-Carlo Tree Search with Policy-Gradient Reinforcement Learning

   • Tobias Graf, Marco Platzner

   Pages 1-11

3. Early Playout Termination in MCTS

   • Richard Lorentz

   Pages 12-19

4. Playout Policy Adaptation for Games

   • Tristan Cazenave

   Pages 20-28

5. Strength Improvement and Analysis for an MCTS-Based Chinese Dark Chess Program

   • Chu-Hsuan Hsueh, I-Chen Wu, Wen-Jie Tseng, Shi-Jim Yen, Jr-Chang Chen

   Pages 29-40

6. LinUCB Applied to Monte-Carlo Tree Search

   • Yusaku Mandai, Tomoyuki Kaneko

   Pages 41-52

7. Adapting Improved Upper Confidence Bounds for Monte-Carlo Tree Search

   • Yun-Ching Liu, Yoshimasa Tsuruoka

   Pages 53-64

8. On Some Random Walk Games with Diffusion Control

   • Ingo Althöfer, Matthias Beckmann, Friedrich Salzer

   Pages 65-75

9. Go Complexities

   • Abdallah Saffidine, Olivier Teytaud, Shi-Jim Yen

   Pages 76-88

10. On Some Evacuation Games with Random Walks

    • Matthias Beckmann

    Pages 89-99

11. Crystallization of Domineering Snowflakes

    • Jos W. H. M. Uiterwijk

    Pages 100-112

12. First Player’s Cannot-Lose Strategies for Cylinder-Infinite-Connect-Four with Widths 2 and 6

    • Yoshiaki Yamaguchi, Todd W. Neller

    Pages 113-121

13. Development of a Program for Playing Progressive Chess

    • Vito Janko, Matej Guid

    Pages 122-134

14. A Comparative Review of Skill Assessment: Performance, Prediction and Profiling

    • Guy Haworth, Tamal Biswas, Ken Regan

    Pages 135-146

15. Boundary Matching for Interactive Sprouts

    • Cameron Browne

    Pages 147-159

16. Draws, Zugzwangs, and PSPACE-Completeness in the Slither Connection Game

    • Édouard Bonnet, Florian Jamain, Abdallah Saffidine

    Pages 160-176

17. Constructing Pin Endgame Databases for the Backgammon Variant Plakoto

    • Nikolaos Papahristou, Ioannis Refanidis

    Pages 177-184

18. Reducing the Seesaw Effect with Deep Proof-Number Search

    • Taichi Ishitobi, Aske Plaat, Hiroyuki Iida, Jaap van den Herik

    Pages 185-197

19. Feature Strength and Parallelization of Sibling Conspiracy Number Search

    • Jakub Pawlewicz, Ryan B. Hayward

    Pages 198-209

20. Parameter-Free Tree Style Pipeline in Asynchronous Parallel Game-Tree Search

    • Shu Yokoyama, Tomoyuki Kaneko, Tetsuro Tanaka

    Pages 210-222

This book constitutes the thoroughly refereed post-conference proceedings of the 14th International Conference on Advances in Computer Games, ACG 2015, held in Leiden, The Netherlands, in July 2015.


The 22 revised full papers presented were carefully reviewed and selected from 34 submissions. The papers cover a wide range of topics such as Monte-Carlo Tree Search and its enhancements;  theoretical aspects and complexity; analysis of game characteristics; search algorithms; and machine learning.

• Machine Learning
• Supervised Learning
• Reinforcement Learning
• optimization
• planning and scheduling
• artificial intelligence
• combinatorial game
• complexity
• distributed computing
• evaluation function
• game tree search
• general game playing
• heuristic function construction
• inductive logic programming
• markov decision process
• parallel search
• random walk
• realtime
• search methodologies
• transfer learning

• Leiden University , Leiden, The Netherlands

  Aske Plaat, Jaap van den Herik, Walter Kosters

Policies and ethics