Losses over Labels: Weakly Supervised Learning via Direct Loss Construction

TL;DR

This paper introduces LoL, a novel weakly supervised learning method that constructs loss functions directly from heuristics, bypassing pseudolabels, and improves performance on text and image classification tasks.

Contribution

The paper proposes a new approach to weak supervision by directly transforming heuristics into loss functions, enabling more informative training and better results.

Findings

LoL outperforms existing weak supervision methods on benchmark tasks.

Incorporating gradient information enhances model performance.

Direct loss construction captures more heuristic information.

Abstract

Owing to the prohibitive costs of generating large amounts of labeled data, programmatic weak supervision is a growing paradigm within machine learning. In this setting, users design heuristics that provide noisy labels for subsets of the data. These weak labels are combined (typically via a graphical model) to form pseudolabels, which are then used to train a downstream model. In this work, we question a foundational premise of the typical weakly supervised learning pipeline: given that the heuristic provides all ``label" information, why do we need to generate pseudolabels at all? Instead, we propose to directly transform the heuristics themselves into corresponding loss functions that penalize differences between our model and the heuristic. By constructing losses directly from the heuristics, we can incorporate more information than is used in the standard weakly supervised pipeline, such as how the heuristics make their decisions, which explicitly informs feature selection during training. We call our method Losses over Labels (LoL) as it creates losses directly from heuristics without going through the intermediate step of a label. We show that LoL improves upon existing weak supervision methods on several benchmark text and image classification tasks and further demonstrate that incorporating gradient information leads to better performance on almost every task.