Slice-based Learning: A Programming Model for Residual Learning in Critical Data Slices

TL;DR

Slice-based Learning introduces a programming model that enhances model performance on critical data subsets by learning slice-specific representations, improving slice and overall accuracy across diverse datasets.

Contribution

The paper proposes Slice-based Learning, a novel programming model that enables models to focus on critical data slices using slice-specific representations and attention mechanisms.

Findings

Up to 19.0 F1 improvement on slices

Up to 4.6 F1 overall improvement

Effective across language, vision, and industrial datasets

Abstract

In real-world machine learning applications, data subsets correspond to especially critical outcomes: vulnerable cyclist detections are safety-critical in an autonomous driving task, and "question" sentences might be important to a dialogue agent's language understanding for product purposes. While machine learning models can achieve high quality performance on coarse-grained metrics like F1-score and overall accuracy, they may underperform on critical subsets---we define these as slices, the key abstraction in our approach. To address slice-level performance, practitioners often train separate "expert" models on slice subsets or use multi-task hard parameter sharing. We propose Slice-based Learning, a new programming model in which the slicing function (SF), a programming interface, specifies critical data subsets for which the model should commit additional capacity. Any model can leverage SFs to learn slice expert representations, which are combined with an attention mechanism to make slice-aware predictions. We show that our approach maintains a parameter-efficient representation while improving over baselines by up to 19.0 F1 on slices and 4.6 F1 overall on datasets spanning language understanding (e.g. SuperGLUE), computer vision, and production-scale industrial systems.