Incremental Deep Neural Network Learning using Classification Confidence Thresholding

TL;DR

This paper introduces a Classification Confidence Threshold method for neural networks to improve incremental learning by efficiently detecting and integrating new classes while minimizing resource use and accuracy loss.

Contribution

The paper proposes a novel confidence threshold approach enabling neural networks to incrementally learn new classes with limited samples and minimal architectural changes.

Findings

Effective detection of unknown classes during testing.

Maintains high accuracy with incremental class addition.

Reduces retraining resources and prevents catastrophic forgetting.

Abstract

Most modern neural networks for classification fail to take into account the concept of the unknown. Trained neural networks are usually tested in an unrealistic scenario with only examples from a closed set of known classes. In an attempt to develop a more realistic model, the concept of working in an open set environment has been introduced. This in turn leads to the concept of incremental learning where a model with its own architecture and initial trained set of data can identify unknown classes during the testing phase and autonomously update itself if evidence of a new class is detected. Some problems that arise in incremental learning are inefficient use of resources to retrain the classifier repeatedly and the decrease of classification accuracy as multiple classes are added over time. This process of instantiating new classes is repeated as many times as necessary, accruing errors. To address these problems, this paper proposes the Classification Confidence Threshold approach to prime neural networks for incremental learning to keep accuracies high by limiting forgetting. A lean method is also used to reduce resources used in the retraining of the neural network. The proposed method is based on the idea that a network is able to incrementally learn a new class even when exposed to a limited number samples associated with the new class. This method can be applied to most existing neural networks with minimal changes to network architecture.