Hierarchical Latent Structure Learning through Online Inference

TL;DR

HOLMES is a new online hierarchical latent structure learning model that balances generalization and discrimination, enabling efficient inference and transfer in sequential data without supervision.

Contribution

Introduces HOLMES, a hierarchical Bayesian model with online inference, combining nested Chinese Restaurant Process and sequential Monte Carlo methods for structure learning.

Findings

HOLMES matches flat models in predictive performance.

HOLMES learns more compact, transferable representations.

HOLMES improves outcome prediction in nested temporal tasks.

Abstract

Learning systems must balance generalization across experiences with discrimination of task-relevant details. Effective learning therefore requires representations that support both. Online latent-cause models support incremental inference but assume flat partitions, whereas hierarchical Bayesian models capture multilevel structure but typically require offline inference. We introduce the Hierarchical Online Learning of Multiscale Experience Structure (HOLMES) model, a computational framework for hierarchical latent structure learning through online inference. HOLMES combines a variation on the nested Chinese Restaurant Process prior with sequential Monte Carlo inference to perform tractable trial-by-trial inference over hierarchical latent representations without explicit supervision over the latent structure. In simulations, HOLMES matched the predictive performance of flat models while learning more compact representations that supported one-shot transfer to higher-level latent categories. In a context-dependent task with nested temporal structure, HOLMES also improved outcome prediction relative to flat models. These results provide a tractable computational framework for discovering hierarchical structure in sequential data.