Context-Aware Drift Detection

TL;DR

This paper introduces a novel context-aware drift detection framework that leverages causal inference and conditional two-sample tests to identify distributional changes in machine learning systems, even when data dependencies exist.

Contribution

It develops a general framework for drift detection based on conditional distribution tests, extending traditional methods to account for context and dependencies.

Findings

Effective in detecting drift in subpopulations of data.

Applicable to large-scale vision datasets like ImageNet.

Insensitive to subpopulation prevalence variations.

Abstract

When monitoring machine learning systems, two-sample tests of homogeneity form the foundation upon which existing approaches to drift detection build. They are used to test for evidence that the distribution underlying recent deployment data differs from that underlying the historical reference data. Often, however, various factors such as time-induced correlation mean that batches of recent deployment data are not expected to form an i.i.d. sample from the historical data distribution. Instead we may wish to test for differences in the distributions conditional on \textit{context} that is permitted to change. To facilitate this we borrow machinery from the causal inference domain to develop a more general drift detection framework built upon a foundation of two-sample tests for conditional distributional treatment effects. We recommend a particular instantiation of the framework based on maximum conditional mean discrepancies. We then provide an empirical study demonstrating its effectiveness for various drift detection problems of practical interest, such as detecting drift in the distributions underlying subpopulations of data in a manner that is insensitive to their respective prevalences. The study additionally demonstrates applicability to ImageNet-scale vision problems.