Strategy Game-Playing with Size-Constrained State Abstraction

2024-08-05

2024 IEEE Conference on Games (CoG)

AI Summary

Strategy-game AI faces the challenge of prohibitively large search spaces. While existing state abstraction methods compress the state space, they lack reliable quality evaluation mechanisms and often require dynamic de-abstraction during search—introducing sensitive hyperparameters. This paper proposes Size-Constrained State Abstraction (SCSA), which enforces an explicit upper bound on the cardinality of each abstract state group, enabling stable, hyperparameter-free abstraction throughout search. SCSA abandons the conventional “abstract-then-deabstract” paradigm, reframing abstraction quality control as an analytically tractable structural constraint. It jointly optimizes abstraction quality via reinforcement learning and game-tree search. Evaluated across three distinct strategy games, SCSA consistently outperforms baseline methods with strong generalization. The implementation is publicly available.

Technology Category

• Search and Optimization: Adversarial Search
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Playing strategy games is a challenging problem for artificial intelligence (AI). One of the major challenges is the large search space due to a diverse set of game components. In recent works, state abstraction has been applied to search-based game AI and has brought significant performance improvements. State abstraction techniques rely on reducing the search space, e.g., by aggregating similar states. However, the application of these abstractions is hindered because the quality of an abstraction is difficult to evaluate. Previous works hence abandon the abstraction in the middle of the search to not bias the search to a local optimum. This mechanism introduces a hyper-parameter to decide the time to abandon the current state abstraction. In this work, we propose a size-constrained state abstraction (SCSA), an approach that limits the maximum number of nodes being grouped together. We found that with SCSA, the abstraction is not required to be abandoned. Our empirical results on 3 strategy games show that the SCSA agent outperforms the previous methods and yields robust performance over different games. Codes are opensourced at https://anonymous.4open.science/r/SCSA-DB44/.

Problem

• Reduces search space in strategy games
• Improves state abstraction quality
• Eliminates need to abandon abstraction

Innovation

• Size-constrained state abstraction
• No need to abandon abstraction
• Outperforms previous methods