Automated Data Slicing for Model Validation:A Big data - AI Integration Approach

TL;DR

This paper introduces Slice Finder, an interactive framework that uses statistical techniques to identify interpretable, problematic data slices in validation datasets, aiding model debugging and validation.

Contribution

It presents a novel method for automatically finding large, interpretable data slices where models perform poorly, enhancing model validation and debugging.

Findings

Effective identification of problematic data slices

Improved model validation on granular data subsets

Application to fairness and fraud detection

Abstract

As machine learning systems become democratized, it becomes increasingly important to help users easily debug their models. However, current data tools are still primitive when it comes to helping users trace model performance problems all the way to the data. We focus on the particular problem of slicing data to identify subsets of the validation data where the model performs poorly. This is an important problem in model validation because the overall model performance can fail to reflect that of the smaller subsets, and slicing allows users to analyze the model performance on a more granular-level. Unlike general techniques (e.g., clustering) that can find arbitrary slices, our goal is to find interpretable slices (which are easier to take action compared to arbitrary subsets) that are problematic and large. We propose Slice Finder, which is an interactive framework for identifying such slices using statistical techniques. Applications include diagnosing model fairness and fraud detection, where identifying slices that are interpretable to humans is crucial. This research is part of a larger trend of Big data and Artificial Intelligence (AI) integration and opens many opportunities for new research.