Incremental Concept Formation over Visual Images Without Catastrophic Forgetting

TL;DR

This paper introduces Cobweb4V, a novel incremental visual classification system inspired by human learning, which effectively learns new concepts over time without catastrophic forgetting, requiring less data and maintaining stable performance.

Contribution

Cobweb4V is a new human-inspired incremental learning method for visual classification that avoids catastrophic forgetting and requires less data than traditional neural networks.

Findings

Effective learning of visual concepts with less data

Stable performance over time without catastrophic forgetting

Good asymptotic behavior in concept formation

Abstract

Deep neural networks have excelled in machine learning, particularly in vision tasks, however, they often suffer from catastrophic forgetting when learning new tasks sequentially. In this work, we introduce Cobweb4V, an alternative to traditional neural network approaches. Cobweb4V is a novel visual classification method that builds on Cobweb, a human like learning system that is inspired by the way humans incrementally learn new concepts over time. In this research, we conduct a comprehensive evaluation, showcasing Cobweb4Vs proficiency in learning visual concepts, requiring less data to achieve effective learning outcomes compared to traditional methods, maintaining stable performance over time, and achieving commendable asymptotic behavior, without catastrophic forgetting effects. These characteristics align with learning strategies in human cognition, positioning Cobweb4V as a promising alternative to neural network approaches.