Multi-scale metrics and self-organizing maps: a computational approach to the structure of sensory maps

TL;DR

This paper proposes a bi-scale metric for self-organizing maps to improve map segmentation, reduce inactive neurons, and adapt to data changes, inspired by neurobiological principles and tested through simulation studies.

Contribution

It introduces a bi-scale metric for SOMs, enhancing their ability to segment data, adapt to changes, and align with neurobiological observations.

Findings

Bi-scale metrics improve neuron activation in SOMs.

Simulations show enhanced map plasticity after data loss.

Tri-scale metrics and variable learning rates offer additional benefits.

Abstract

This paper introduces the concept of a bi-scale metric for use in the cooperative phase of the self-organizing map (SOM) algorithm. Use of a bi-scale metric allows segmentation of the map into a number of regions, corresponding to anticipated cluster structure in the data. Such a situation occurs, for example, in the somatotopic maps which inspired the SOM algo- rithm, where clusters of data may correspond to body surface regions whose general structure is known. When a bi-scale metric is appropriately applied, issues with map neurons that are not activated by any point in the training data are reduced or eliminated. The paper also presents results of simulation studies on the plasticity of bi-scale metric maps when they are retrained af- ter loss of groups of map neurons or after changes in training data (such as would occur in a somatotopic map when a body surface region like a finger is lost/removed). The paper further considers situations where tri-scale met- rics may be useful, and an alternative approach suggested by neurobiology, where some map regions adapt more slowly to stimuli because they have a lower learning rate parameter.