Learning to Imitate with Less: Efficient Individual Behavior Modeling in Chess

2025-07-29

AI Summary

Existing personalized AI decision modeling—e.g., in chess—relies heavily on large-scale individual game data (typically ≥5,000 games), limiting generalizability to new or data-scarce users. Method: We propose Maia4All, the first prototype-augmented two-stage framework: (1) a behavior enrichment stage that fuses population-level strategy prototypes to construct structured prior knowledge; and (2) a democratization stage that initializes user-specific embeddings hierarchically by skill level and fine-tunes them efficiently. Contribution/Results: Maia4All achieves high-fidelity reconstruction of individual move preferences and behavioral profiles using only 20 games—improving data efficiency by 250× over prior methods. Extensive experiments demonstrate significant gains over state-of-the-art baselines in chess modeling and successful transfer to LLM personalization tasks. The framework establishes a scalable, interpretable paradigm for low-resource personalized AI, enabling robust adaptation with minimal per-user supervision.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Multiagent Systems: Mechanism Design
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search

Abstract

As humans seek to collaborate with, learn from, and better understand artificial intelligence systems, developing AIs that can accurately emulate individual decision-making becomes increasingly important. Chess, a long-standing AI benchmark with precise skill measurement, offers an ideal testbed for human-AI alignment. However, existing approaches to modeling human behavior require prohibitively large amounts of data from each individual, making them impractical for new or sparsely represented users. In this work, we introduce Maia4All, a framework designed to learn and adapt to individual decision-making styles efficiently, even with limited data. Maia4All achieves this through a two-stage optimization process: (1) an enrichment step, which bridges population and individual-level human behavior modeling with a prototype-enriched model, and (2) a democratization step, which leverages ability levels or user prototypes to initialize and refine individual embeddings with minimal data. Our experimental results show that Maia4All can accurately predict individual moves and profile behavioral patterns with high fidelity, establishing a new standard for personalized human-like AI behavior modeling in chess. Maia4All achieves individual human behavior modeling in chess with only 20 games, compared to the 5,000 games required previously, representing a significant improvement in data efficiency. Our work provides an example of how population AI systems can flexibly adapt to individual users using a prototype-enriched model as a bridge. This approach extends beyond chess, as shown in our case study on idiosyncratic LLMs, highlighting its potential for broader applications in personalized AI adaptation.

Problem

• Efficiently model individual chess decisions with limited data
• Reduce required games from 5000 to 20 per player
• Adapt AI systems to personalized human behavior patterns

Innovation

• Two-stage optimization for efficient modeling
• Prototype-enriched model bridges population-individual gaps
• Democratization step refines embeddings with minimal data