Continual General Chunking Problem and SyncMap

TL;DR

This paper introduces SyncMap, an unsupervised, adaptive algorithm for continual chunking of sequences that outperforms or matches existing methods across diverse structural scenarios, highlighting the power of self-organization.

Contribution

The paper presents SyncMap, a novel dynamic map algorithm capable of learning and adapting to various chunking structures and their continual changes without relying on loss functions.

Findings

SyncMap learns near optimal solutions across diverse structures.

SyncMap outperforms or ties with existing algorithms in 66% of scenarios.

SyncMap demonstrates the effectiveness of self-organization in sequence learning.

Abstract

Humans possess an inherent ability to chunk sequences into their constituent parts. In fact, this ability is thought to bootstrap language skills and learning of image patterns which might be a key to a more animal-like type of intelligence. Here, we propose a continual generalization of the chunking problem (an unsupervised problem), encompassing fixed and probabilistic chunks, discovery of temporal and causal structures and their continual variations. Additionally, we propose an algorithm called SyncMap that can learn and adapt to changes in the problem by creating a dynamic map which preserves the correlation between variables. Results of SyncMap suggest that the proposed algorithm learn near optimal solutions, despite the presence of many types of structures and their continual variation. When compared to Word2vec, PARSER and MRIL, SyncMap surpasses or ties with the best algorithm on $66\%$ of the scenarios while being the second best in the remaining $34\%$. SyncMap's model-free simple dynamics and the absence of loss functions reveal that, perhaps surprisingly, much can be done with self-organization alone. Code available at https://github.com/zweifel/SyncMap.