Saved in:
Bibliographic Details
Main Authors: Urbanik, Igor, Gajewski, Paweł
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.10680
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918394444382208
author Urbanik, Igor
Gajewski, Paweł
author_facet Urbanik, Igor
Gajewski, Paweł
contents Continual learning poses a fundamental challenge for neural systems, which often suffer from catastrophic forgetting when exposed to sequential tasks. Self-Organizing Maps (SOMs), despite their interpretability and efficiency, are not immune to this issue. In this paper, we introduce Saturation Self-Organizing Maps (SatSOM)-an extension of SOMs designed to improve knowledge retention in continual learning scenarios. SatSOM incorporates a novel saturation mechanism that gradually reduces the learning rate and neighborhood radius of neurons as they accumulate information. This effectively freezes well-trained neurons and redirects learning to underutilized areas of the map.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10680
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SatSOM: Saturation Self-Organizing Maps for Continual Learning
Urbanik, Igor
Gajewski, Paweł
Machine Learning
Artificial Intelligence
Continual learning poses a fundamental challenge for neural systems, which often suffer from catastrophic forgetting when exposed to sequential tasks. Self-Organizing Maps (SOMs), despite their interpretability and efficiency, are not immune to this issue. In this paper, we introduce Saturation Self-Organizing Maps (SatSOM)-an extension of SOMs designed to improve knowledge retention in continual learning scenarios. SatSOM incorporates a novel saturation mechanism that gradually reduces the learning rate and neighborhood radius of neurons as they accumulate information. This effectively freezes well-trained neurons and redirects learning to underutilized areas of the map.
title SatSOM: Saturation Self-Organizing Maps for Continual Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.10680