Multi-Instance Partial-Label Learning: Towards Exploiting Dual Inexact Supervision

TL;DR

This paper introduces MIPLGP, a novel Gaussian process-based algorithm designed to effectively learn from complex data where each sample has multiple instances and a candidate label set containing one true label and some false positives.

Contribution

The paper formalizes multi-instance partial-label learning (MIPL) and proposes MIPLGP, a tailored algorithm that disambiguates labels and models data more effectively than existing methods.

Findings

MIPLGP outperforms existing algorithms on various datasets.

The proposed method effectively disambiguates candidate label sets.

Experimental results demonstrate superior accuracy and robustness.

Abstract

Weakly supervised machine learning algorithms are able to learn from ambiguous samples or labels, e.g., multi-instance learning or partial-label learning. However, in some real-world tasks, each training sample is associated with not only multiple instances but also a candidate label set that contains one ground-truth label and some false positive labels. Specifically, at least one instance pertains to the ground-truth label while no instance belongs to the false positive labels. In this paper, we formalize such problems as multi-instance partial-label learning (MIPL). Existing multi-instance learning algorithms and partial-label learning algorithms are suboptimal for solving MIPL problems since the former fail to disambiguate a candidate label set, and the latter cannot handle a multi-instance bag. To address these issues, a tailored algorithm named MIPLGP, i.e., Multi-Instance Partial-Label learning with Gaussian Processes, is proposed. MIPLGP first assigns each instance with a candidate label set in an augmented label space, then transforms the candidate label set into a logarithmic space to yield the disambiguated and continuous labels via an exclusive disambiguation strategy, and last induces a model based on the Gaussian processes. Experimental results on various datasets validate that MIPLGP is superior to well-established multi-instance learning and partial-label learning algorithms for solving MIPL problems. Our code and datasets will be made publicly available.