Unsupervised Concept Drift Detection based on Parallel Activations of Neural Network

TL;DR

This paper introduces an unsupervised concept drift detection method using parallel activations from an untrained neural network, addressing the challenge of label scarcity in streaming data environments.

Contribution

It presents a novel unsupervised drift detector based on neural network activations, eliminating the need for labeled data and demonstrating competitive performance.

Findings

Competitive detection accuracy compared to state-of-the-art methods

Effective in scenarios with limited or delayed labels

Applicable to real-time streaming data environments

Abstract

Practical applications of artificial intelligence increasingly often have to deal with the streaming properties of real data, which, considering the time factor, are subject to phenomena such as periodicity and more or less chaotic degeneration - resulting directly in the concept drifts. The modern concept drift detectors almost always assume immediate access to labels, which due to their cost, limited availability and possible delay has been shown to be unrealistic. This work proposes an unsupervised Parallel Activations Drift Detector, utilizing the outputs of an untrained neural network, presenting its key design elements, intuitions about processing properties, and a pool of computer experiments demonstrating its competitiveness with state-of-the-art methods.