Multi-Attention Multiple Instance Learning

TL;DR

The paper introduces MAMIL, a multi-attention approach for MIL that considers neighboring patches, uses diverse attention modules for feature representation, and improves classification accuracy and interpretability.

Contribution

It proposes a novel multi-attention MIL method that incorporates neighboring patches and diverse feature representations for enhanced classification.

Findings

Effective in processing different patch types

Provides diverse feature representations of patches

Achieves accurate classification results

Abstract

A new multi-attention based method for solving the MIL problem (MAMIL), which takes into account the neighboring patches or instances of each analyzed patch in a bag, is proposed. In the method, one of the attention modules takes into account adjacent patches or instances, several attention modules are used to get a diverse feature representation of patches, and one attention module is used to unite different feature representations to provide an accurate classification of each patch (instance) and the whole bag. Due to MAMIL, a combined representation of patches and their neighbors in the form of embeddings of a small dimensionality for simple classification is realized. Moreover, different types of patches are efficiently processed, and a diverse feature representation of patches in a bag by using several attention modules is implemented. A simple approach for explaining the classification predictions of patches is proposed. Numerical experiments with various datasets illustrate the proposed method.