From Dionysius Emerges Apollo -- Learning Patterns and Abstractions from Perceptual Sequences
🤖 AI Summary
This paper addresses how high-dimensional perceptual sequences are rapidly decomposed into reusable hierarchical structures to support learning, generalization, and prediction. Inspired by human “chunking” behavior, we propose a nonparametric hierarchical latent-variable model grounded in rational pattern discovery, unifying sequence segmentation, hierarchical abstraction, cross-task transfer, and symbolic representation within a single computational framework. Our method integrates unsupervised pattern discovery, nonparametric Bayesian modeling, and behavioral experimentation, enabling end-to-end hierarchical learning on both univariate and multivariate sequences. It faithfully reproduces key human chunking phenomena, supports zero-shot transfer, enables cognitive process simulation, and achieves large-language-model-level performance in abstract motif identification. The core contribution is the first computational account demonstrating how chunking and abstraction jointly drive emergent structural organization—thereby bridging cognitive science and modern sequence modeling.
Application Category
📝 Abstract
Cognition swiftly breaks high-dimensional sensory streams into familiar parts and uncovers their relations. Why do structures emerge, and how do they enable learning, generalization, and prediction? What computational principles underlie this core aspect of perception and intelligence? A sensory stream, simplified, is a one-dimensional sequence. In learning such sequences, we naturally segment them into parts -- a process known as chunking. In the first project, I investigated factors influencing chunking in a serial reaction time task and showed that humans adapt to underlying chunks while balancing speed and accuracy. Building on this, I developed models that learn chunks and parse sequences chunk by chunk. Normatively, I proposed chunking as a rational strategy for discovering recurring patterns and nested hierarchies, enabling efficient sequence factorization. Learned chunks serve as reusable primitives for transfer, composition, and mental simulation -- letting the model compose the new from the known. I demonstrated this model's ability to learn hierarchies in single and multi-dimensional sequences and highlighted its utility for unsupervised pattern discovery. The second part moves from concrete to abstract sequences. I taxonomized abstract motifs and examined their role in sequence memory. Behavioral evidence suggests that humans exploit pattern redundancies for compression and transfer. I proposed a non-parametric hierarchical variable model that learns both chunks and abstract variables, uncovering invariant symbolic patterns. I showed its similarity to human learning and compared it to large language models. Taken together, this thesis suggests that chunking and abstraction as simple computational principles enable structured knowledge acquisition in hierarchically organized sequences, from simple to complex, concrete to abstract.
Problem
Research questions and friction points this paper is trying to address.
Investigates chunking in sequence learning and its computational principles.
Develops models for learning chunks and parsing sequences hierarchically.
Explores abstraction in sequences and its role in memory and transfer.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Chunking enables efficient sequence factorization.
Non-parametric model learns chunks and abstract variables.
Hierarchical learning supports structured knowledge acquisition.
🔎 Similar Papers
No similar papers found.
