Anomaly Attribution with Likelihood Compensation

TL;DR

This paper introduces a likelihood compensation method to explain anomalous predictions of black-box regression models by assigning responsibility scores to input variables, validated on building energy data.

Contribution

It presents the first principled framework for computing responsibility scores for real-valued model deviations in anomaly attribution.

Findings

Effective in real-world building energy prediction

Provides interpretable responsibility scores

Confirmed utility through expert feedback

Abstract

This paper addresses the task of explaining anomalous predictions of a black-box regression model. When using a black-box model, such as one to predict building energy consumption from many sensor measurements, we often have a situation where some observed samples may significantly deviate from their prediction. It may be due to a sub-optimal black-box model, or simply because those samples are outliers. In either case, one would ideally want to compute a ``responsibility score'' indicative of the extent to which an input variable is responsible for the anomalous output. In this work, we formalize this task as a statistical inverse problem: Given model deviation from the expected value, infer the responsibility score of each of the input variables. We propose a new method called likelihood compensation (LC), which is founded on the likelihood principle and computes a correction to each input variable. To the best of our knowledge, this is the first principled framework that computes a responsibility score for real valued anomalous model deviations. We apply our approach to a real-world building energy prediction task and confirm its utility based on expert feedback.