Analogy making as amortised model construction

2025-07-22

AI Summary

How can agents rapidly construct accurate, reusable internal models under resource constraints? Method: We propose an analogy-driven experience transfer framework that formalizes analogy as a partial homomorphism between Markov decision processes (MDPs), enabling the composition and reuse of abstract modules. This framework tightly couples modular representation learning with computational cognitive modeling. Contribution/Results: Our approach significantly amortizes the costs of model construction and planning while supporting flexible cross-domain transfer of policies and representations. Experiments demonstrate that analogy-guided module reuse effectively reduces modeling complexity, accelerates adaptation to novel environments, and improves planning efficiency. The framework establishes a new paradigm for general-purpose modeling in resource-constrained agents.

Technology Category

• Cognitive Modeling & Cognitive Systems: Analogy
• Multiagent Systems: Modeling other Agents
• Planning, Routing, and Scheduling: Planning with Markov Models (MDPs, POMDPs)

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Humans flexibly construct internal models to navigate novel situations. To be useful, these internal models must be sufficiently faithful to the environment that resource-limited planning leads to adequate outcomes; equally, they must be tractable to construct in the first place. We argue that analogy plays a central role in these processes, enabling agents to reuse solution-relevant structure from past experiences and amortise the computational costs of both model construction (construal) and planning. Formalising analogies as partial homomorphisms between Markov decision processes, we sketch a framework in which abstract modules, derived from previous construals, serve as composable building blocks for new ones. This modular reuse allows for flexible adaptation of policies and representations across domains with shared structural essence.

Problem

• How humans construct internal models for novel situations
• Role of analogy in reusing past experiences efficiently
• Modular reuse of abstract structures across domains

Innovation

• Analogy enables reuse of past solution structures
• Formalizes analogies as partial MDP homomorphisms
• Modular abstract blocks adapt policies across domains