Debugging Machine Learning Tasks

TL;DR

This paper introduces Psi, an automated causality-based tool for debugging data errors in machine learning classification tasks, addressing a gap in existing debugging tools for data rather than code.

Contribution

It proposes a novel method using Pearl's causation theory and probabilistic programming to identify root causes of misclassifications due to data errors in ML tasks.

Findings

Psi effectively identifies root causes of data errors.

The method leverages Pearl's PS metric for causality analysis.

Experimental results show Psi's utility on real datasets.

Abstract

Unlike traditional programs (such as operating systems or word processors) which have large amounts of code, machine learning tasks use programs with relatively small amounts of code (written in machine learning libraries), but voluminous amounts of data. Just like developers of traditional programs debug errors in their code, developers of machine learning tasks debug and fix errors in their data. However, algorithms and tools for debugging and fixing errors in data are less common, when compared to their counterparts for detecting and fixing errors in code. In this paper, we consider classification tasks where errors in training data lead to misclassifications in test points, and propose an automated method to find the root causes of such misclassifications. Our root cause analysis is based on Pearl's theory of causation, and uses Pearl's PS (Probability of Sufficiency) as a scoring metric. Our implementation, Psi, encodes the computation of PS as a probabilistic program, and uses recent work on probabilistic programs and transformations on probabilistic programs (along with gray-box models of machine learning algorithms) to efficiently compute PS. Psi is able to identify root causes of data errors in interesting data sets.