Cover image for General Video Game Artificial Intelligence

General Video Game Artificial Intelligence

Author Diego Pérez Liébana , Simon M. Lucas , Raluca D. Gaina
Release Date: 2019/10/01
Topic:
0%
96
Chapters
0-1
Hours read
0k
Total Words
    Start Reading Now    
       Add to Wishlist       
View table of contents

Book Description

Research on general video game playing aims at designing agents or content generators that can perform well in multiple video games, possibly without knowing the game in advance and with little to no specific domain knowledge.

The general video game AI framework and competition propose a challenge in which researchers can test their favorite AI methods with a potentially infinite number of games created using the Video Game Description Language. The open-source framework has been used since 2014 for running a challenge. Competitors around the globe submit their best approaches that aim to generalize well across games. Additionally, the framework has been used in AI modules by many higher-education institutions as assignments, or as proposed projects for final year (undergraduate and Master's) students and Ph.D. candidates.

The present book, written by the developers and organizers of the framework, presents the most interesting highlights of the research performed by the authors during these years in this domain. It showcases work on methods to play the games, generators of content, and video game optimization. It also outlines potential further work in an area that offers multiple research directions for the future.

Table of Contents

  1. Preface
  2. Acknowledgments
  3. Introduction
    1. A Historical View: From Chess to GVGAI
    2. GVGAI in Education
    3. GVGAI and the Games Industry
  4. VGDL and the GVGAI Framework
    1. Introduction
    2. The Video Game Description Language
      1. Continuous Physics
      2. 2-Player Games
      3. The Future of VGDL
    3. The General Video Game AI Framework
      1. API for 2-Player Games
      2. GVGAI as a Multi-Track Challenge
    4. The GVGAI Competition
    5. Exercises
      1. Running GVGAI
      2. VGDL
      3. Submit to the GVGAI competition server
  5. Planning in GVGAI
    1. Introduction
    2. Monte Carlo Tree Search
    3. Knowledge-Based Fast Evolutionary MCTS
      1. Fast Evolution in MCTS
      2. Learning Domain Knowledge
      3. Experimental Work
    4. Multi-Objective MCTS for GVGAI
      1. Multi-Objective Optimization
      2. Multi-Objective MCTS
      3. Experimental Work
    5. Rolling Horizon Evolutionary Algorithms
      1. Vanilla RHEA
      2. RHEA Enhancements
    6. Exercises
      1. Monte Carlo Tree Search
      2. Rolling Horizon Evolutionary Algorithms
  6. Frontiers of GVGAI Planning
    1. Introduction
    2. State of the Art in GVGAI Planning
      1. OLETS (adrienctx)
      2. ToVo2
      3. YOLOBOT
    3. Current Problems in GVGAI Planning
    4. General Win Prediction in GVGAI
      1. Game Playing Agents and Features
      2. Predictive Models (1/2)
      3. Predictive Models (2/2)
    5. Exercises
      1. Current Problems in GVGAI Planning
      2. General Win Prediction
  7. Learning in GVGAI
    1. Challenges of Learning in GVGAI
    2. Framework
      1. GVGAI Learning Framework
      2. GVGAI Learning Environment
      3. GVGAI Gym Environment
      4. Comparing to Other Learning Frameworks
    3. GVGAI Learning Competitions
      1. Competition Using the GVGAI Learning Environment
      2. Competition Using the GVGAI Gym
    4. Competition Entries
      1. Random Agent (Sample Agents)
      2. DRL Algorithms (Sample Agents)
      3. Multi-Armed Bandit Algorithm
      4. Sarsa
      5. Q-Learning
      6. Tree Search Methods
      7. Other Learning Agents
      8. Discussion
    5. Summary
    6. Exercises
      1. Download and Installation
      2. Play a Game Randomly
      3. Train a More Sophisticated Agent
      4. Create Your Own Agents
  8. Procedural Content Generation in GVGAI
    1. Level Generation in GVGAI
      1. Sample Generators
      2. Competition and Other Generators
      3. Discussion
    2. Rule Generation in GVGAI
      1. Sample Generators
      2. Other Generators
      3. Discussion
    3. Exercises
  9. Automatic General Game Tuning
    1. Introduction
      1. Previous Work
    2. GVGAI Parameterization
    3. Evolving Games for Different Agents
      1. The N-Tuple Bandit Evolutionary Algorithm
      2. Variants of Aliens for Agents with Different Look-Aheads
    4. Modeling Player Experience
      1. Designing the Search Space
      2. Evolving Games for Player Experience (1/2)
      3. Evolving Games for Player Experience (2/2)
    5. Exercises
      1. Parameterizing VGDL Games
      2. Optimize VGDL Games
  10. GVGAI without VGDL
    1. Introduction
    2. Implementation Principles
      1. Copying the Game State
      2. Advancing the Game State
    3. Interfacing
      1. Running GVGAI Agents on the Java Games
      2. Running New Agents on the VGDL Games
    4. Sample Java Games
      1. Asteroids
      2. Fast Planet Wars
      3. Cave Swing
      4. Speed
    5. Conclusions
    6. Exercises
      1. Running the Demos
      2. RHEA Exercise
      3. Agent Comparison Exercise
      4. Game Tuning Exercise
  11. GVGAI: What's Next?
  12. Bibliography (1/3)
  13. Bibliography (2/3)
  14. Bibliography (3/3)
  15. Authors' Biographies
  16. Blank Page (1/3)
  17. Blank Page (2/3)
  18. Blank Page (3/3)