A Mathematical Formalization of Hierarchical Temporal Memory's Spatial Pooler
This work formalizes a component of an emerging algorithm for spatiotemporal data prediction, but it is incremental as it builds on existing HTM concepts without major breakthroughs.
The authors tackled the lack of a comprehensive mathematical framework for the spatial pooler in hierarchical temporal memory by proposing a unifying framework, including a maximum likelihood estimator for permanence updates, and verified that it can be used for feature learning with proper parameterizations.
Hierarchical temporal memory (HTM) is an emerging machine learning algorithm, with the potential to provide a means to perform predictions on spatiotemporal data. The algorithm, inspired by the neocortex, currently does not have a comprehensive mathematical framework. This work brings together all aspects of the spatial pooler (SP), a critical learning component in HTM, under a single unifying framework. The primary learning mechanism is explored, where a maximum likelihood estimator for determining the degree of permanence update is proposed. The boosting mechanisms are studied and found to be only relevant during the initial few iterations of the network. Observations are made relating HTM to well-known algorithms such as competitive learning and attribute bagging. Methods are provided for using the SP for classification as well as dimensionality reduction. Empirical evidence verifies that given the proper parameterizations, the SP may be used for feature learning.