Learning from Similarity-Confidence and Confidence-Difference

TL;DR

This paper introduces a novel weakly supervised learning framework that combines similarity-confidence and confidence-difference signals, with theoretical guarantees and improved performance over existing methods.

Contribution

It proposes a new WSL method leveraging multiple weak supervision signals and derives unbiased risk estimators with proven optimal convergence rates.

Findings

Outperforms existing baselines across various settings.

Provides theoretical analysis on robustness to label noise.

Introduces risk correction to prevent overfitting.

Abstract

In practical machine learning applications, it is often challenging to assign accurate labels to data, and increasing the number of labeled instances is often limited. In such cases, Weakly Supervised Learning (WSL), which enables training with incomplete or imprecise supervision, provides a practical and effective solution. However, most existing WSL methods focus on leveraging a single type of weak supervision. In this paper, we propose a novel WSL framework that leverages complementary weak supervision signals from multiple relational perspectives, which can be especially valuable when labeled data is limited. Specifically, we introduce SconfConfDiff Classification, a method that integrates two distinct forms of weaklabels: similarity-confidence and confidence-difference, which are assigned to unlabeled data pairs. To implement this method, we derive two types of unbiased risk estimators for classification: one based on a convex combination of existing estimators, and another newly designed by modeling the interaction between two weak labels. We prove that both estimators achieve optimal convergence rates with respect to estimation error bounds. Furthermore, we introduce a risk correction approach to mitigate overfitting caused by negative empirical risk, and provide theoretical analysis on the robustness of the proposed method against inaccurate class prior probability and label noise. Experimental results demonstrate that the proposed method consistently outperforms existing baselines across a variety of settings.