Disambiguated Attention Embedding for Multi-Instance Partial-Label Learning

TL;DR

This paper introduces DEMIPL, a novel method for multi-instance partial-label learning that uses attention-based embedding and disambiguation strategies to improve label prediction accuracy, validated on benchmark and real-world datasets.

Contribution

The paper proposes DEMIPL, an innovative embedding and disambiguation approach for MIPL that outperforms existing methods and introduces a new colorectal cancer classification dataset.

Findings

DEMIPL achieves superior accuracy on benchmark datasets.

DEMIPL outperforms existing MIPL and partial-label learning methods.

The real-world colorectal cancer dataset demonstrates DEMIPL's practical effectiveness.

Abstract

In many real-world tasks, the concerned objects can be represented as a multi-instance bag associated with a candidate label set, which consists of one ground-truth label and several false positive labels. Multi-instance partial-label learning (MIPL) is a learning paradigm to deal with such tasks and has achieved favorable performances. Existing MIPL approach follows the instance-space paradigm by assigning augmented candidate label sets of bags to each instance and aggregating bag-level labels from instance-level labels. However, this scheme may be suboptimal as global bag-level information is ignored and the predicted labels of bags are sensitive to predictions of negative instances. In this paper, we study an alternative scheme where a multi-instance bag is embedded into a single vector representation. Accordingly, an intuitive algorithm named DEMIPL, i.e., Disambiguated attention Embedding for Multi-Instance Partial-Label learning, is proposed. DEMIPL employs a disambiguation attention mechanism to aggregate a multi-instance bag into a single vector representation, followed by a momentum-based disambiguation strategy to identify the ground-truth label from the candidate label set. Furthermore, we introduce a real-world MIPL dataset for colorectal cancer classification. Experimental results on benchmark and real-world datasets validate the superiority of DEMIPL against the compared MIPL and partial-label learning approaches.