Unsupervised Concept Drift Detection based on Parallel Activations of Neural Network
This addresses the challenge of concept drift detection in real-world AI applications where labels are costly or delayed, offering an unsupervised solution.
The paper tackles the problem of detecting concept drift in streaming data without requiring labeled data, proposing an unsupervised method that uses outputs from an untrained neural network and shows competitiveness with state-of-the-art methods in experiments.
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.