Calibratable Disambiguation Loss for Multi-Instance Partial-Label Learning

TL;DR

This paper introduces a calibratable disambiguation loss (CDL) for multi-instance partial-label learning that improves both classification accuracy and calibration reliability, addressing a key weakness of existing methods.

Contribution

The paper proposes a novel plug-and-play CDL that enhances calibration in MIPL, with theoretical analysis and superior empirical performance over traditional disambiguation losses.

Findings

CDL improves calibration and accuracy in MIPL tasks.

Theoretical analysis shows CDL's lower bound and regularization benefits.

Experimental results confirm CDL's effectiveness on benchmark datasets.

Abstract

Multi-instance partial-label learning (MIPL) is a weakly supervised framework that extends the principles of multi-instance learning (MIL) and partial-label learning (PLL) to address the challenges of inexact supervision in both instance and label spaces. However, existing MIPL approaches often suffer from poor calibration, undermining classifier reliability. In this work, we propose a plug-and-play calibratable disambiguation loss (CDL) that simultaneously improves classification accuracy and calibration performance. The loss has two instantiations: the first one calibrates predictions based on probabilities from the candidate label set, while the second one integrates probabilities from both candidate and non-candidate label sets. The proposed CDL can be seamlessly incorporated into existing MIPL and PLL frameworks. We provide a theoretical analysis that establishes the lower bound and regularization properties of CDL, demonstrating its superiority over conventional disambiguation losses. Experimental results on benchmark and real-world datasets confirm that our CDL significantly enhances both classification and calibration performance.