Latent learning: episodic memory complements parametric learning by enabling flexible reuse of experiences

2025-09-19

AI Summary

Machine learning systems often suffer from limited generalization due to the absence of latent learning—the capacity to acquire task-irrelevant yet transferable implicit knowledge beneficial for future tasks. This work identifies an inherent limitation of parametric learning and, for the first time, introduces the cognitive science concept of latent learning into machine learning, proposing a novel paradigm that integrates episodic memory with in-context learning. Methodologically, we design an episodic memory module equipped with an oracle-based retrieval mechanism, enabling dynamic storage, efficient retrieval, and cross-task reuse of historical experiences. Experiments on language modeling and agent navigation demonstrate substantial improvements in cross-task generalization and mitigate data inefficiency. Our core contributions are threefold: (1) establishing the lack of latent learning as a fundamental generalization bottleneck; (2) introducing a retrieval-augmented episodic memory mechanism; and (3) empirically validating its effectiveness as a complementary alternative to parametric learning.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Search and Optimization: Learning to Search
• Multiagent Systems: Multiagent Learning

Application Category

• 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: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

When do machine learning systems fail to generalize, and what mechanisms could improve their generalization? Here, we draw inspiration from cognitive science to argue that one weakness of machine learning systems is their failure to exhibit latent learning -- learning information that is not relevant to the task at hand, but that might be useful in a future task. We show how this perspective links failures ranging from the reversal curse in language modeling to new findings on agent-based navigation. We then highlight how cognitive science points to episodic memory as a potential part of the solution to these issues. Correspondingly, we show that a system with an oracle retrieval mechanism can use learning experiences more flexibly to generalize better across many of these challenges. We also identify some of the essential components for effectively using retrieval, including the importance of within-example in-context learning for acquiring the ability to use information across retrieved examples. In summary, our results illustrate one possible contributor to the relative data inefficiency of current machine learning systems compared to natural intelligence, and help to understand how retrieval methods can complement parametric learning to improve generalization.

Problem

• Addresses machine learning failures in generalization
• Proposes episodic memory to enable flexible experience reuse
• Identifies retrieval mechanisms to improve data efficiency

Innovation

• Episodic memory enables flexible experience reuse
• Oracle retrieval mechanism improves generalization
• Within-example in-context learning facilitates retrieval