mlsanityData to 14 Oct 2026

Multi-Attention Multiple Instance Learning

arXiv 2112.06071 ↗First posted 11 Dec 2021cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 21 of its 67 references that are in the map. First posted on arXiv 11 Dec 2021.

Builds on

21 of 67 references.

References · 21

  1. 2017Attention Is All You Needcited by 60,361
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  3. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
  4. 2019Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AIcited by 1,552
  5. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  6. 2015Effective Approaches to Attention-based Neural Machine Translationcited by 1,026
  7. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 478
  8. 2021Interpretable Machine Learning: Fundamental Principles and 10 Grand Challengescited by 367
  9. 2018Not-so-supervised: a survey of semi-supervised, multi-instance, and transfer learning in medical image analysiscited by 179
  10. 2020Whole Slide Images based Cancer Survival Prediction using Attention Guided Deep Multiple Instance Learning Networkscited by 178
  11. 2019Deep neural network models for computational histopathology: A surveycited by 165
  12. 2016Multiple Instance Learning: A Survey of Problem Characteristics and Applicationscited by 164
9 more
  1. 2019An Attentive Survey of Attention Modelscited by 124
  2. 2016Revisiting Multiple Instance Neural Networkscited by 119
  3. 2021Interpretable Deep Learning: Interpretation, Interpretability, Trustworthiness, and Beyondcited by 118
  4. 2021Dive into Deep Learningcited by 111
  5. 2015Classifying and Segmenting Microscopy Images Using Convolutional Multiple Instance Learningcited by 38
  6. 2019Resolving challenges in deep learning-based analyses of histopathological images using explanation methodscited by 32
  7. 2021MHAttnSurv: Multi-Head Attention for Survival Prediction Using Whole-Slide Pathology Imagescited by 15
  8. 2021Nested Multiple Instance Learning with Attention Mechanismscited by 7
  9. 2016Multiple Instance Learning Convolutional Neural Networks for Object Recognitioncited by 6

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 4.

6 papers.

  1. 2022Magnification Prior: A Self-Supervised Method for Learning Representations on Breast Cancer Histopathological Imagescited by 9
  2. 2023Towards Hierarchical Regional Transformer-based Multiple Instance Learningcited by 4
  3. 2022Interpreting Vulnerabilities of Multi-Instance Learning to Adversarial Perturbationscited by 3
  4. 2023Data-driven Knowledge Fusion for Deep Multi-instance Learningcited by 3
  5. 2023ProMIL: Probabilistic Multiple Instance Learning for Medical Imagingcited by 3
  6. 2023Multiple Instance Learning with Trainable Decision Tree Ensemblescited by 0