Concept Drift Detection from Multi-Class Imbalanced Data Streams

TL;DR

This paper introduces a novel trainable concept drift detector based on Restricted Boltzmann Machines, designed to handle multi-class imbalanced data streams by monitoring classes independently and adapting to evolving class roles.

Contribution

It proposes a new taxonomy of challenges in multi-class imbalanced concept drift detection and develops a skew-insensitive, trainable detector capable of handling local drifts and changing class distributions.

Findings

High efficacy demonstrated through extensive experiments

Effective handling of local concept drifts in minority classes

Robust detection across evolving class imbalance ratios

Abstract

Continual learning from data streams is among the most important topics in contemporary machine learning. One of the biggest challenges in this domain lies in creating algorithms that can continuously adapt to arriving data. However, previously learned knowledge may become outdated, as streams evolve over time. This phenomenon is known as concept drift and must be detected to facilitate efficient adaptation of the learning model. While there exists a plethora of drift detectors, all of them assume that we are dealing with roughly balanced classes. In the case of imbalanced data streams, those detectors will be biased towards the majority classes, ignoring changes happening in the minority ones. Furthermore, class imbalance may evolve over time and classes may change their roles (majority becoming minority and vice versa). This is especially challenging in the multi-class setting, where relationships among classes become complex. In this paper, we propose a detailed taxonomy of challenges posed by concept drift in multi-class imbalanced data streams, as well as a novel trainable concept drift detector based on Restricted Boltzmann Machine. It is capable of monitoring multiple classes at once and using reconstruction error to detect changes in each of them independently. Our detector utilizes a skew-insensitive loss function that allows it to handle multiple imbalanced distributions. Due to its trainable nature, it is capable of following changes in a stream and evolving class roles, as well as it can deal with local concept drift occurring in minority classes. Extensive experimental study on multi-class drifting data streams, enriched with a detailed analysis of the impact of local drifts and changing imbalance ratios, confirms the high efficacy of our approach.