mlsanityData to 7 Oct 2026

Interventional Multi-Instance Learning with Deconfounded Instance-Level Prediction

arXiv 2204.09204 ↗First posted 20 Apr 2022cs.CV cs.LGcited by 13

Not on the map. Why?

Abstract (arXiv)

When applying multi-instance learning (MIL) to make predictions for bags of instances, the prediction accuracy of an instance often depends on not only the instance itself but also its context in the corresponding bag. From the viewpoint of causal inference, such bag contextual prior works as a confounder and may result in model robustness and interpretability issues. Focusing on this problem, we propose a novel interventional multi-instance learning (IMIL) framework to achieve deconfounded instance-level prediction. Unlike traditional likelihood-based strategies, we design an Expectation-Maximization (EM) algorithm based on causal intervention, providing a robust instance selection in the training phase and suppressing the bias caused by the bag contextual prior. Experiments on pathological image analysis demonstrate that our IMIL method substantially reduces false positives and outperforms state-of-the-art MIL methods.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 32 of its 53 references that are in the map. First posted on arXiv 20 Apr 2022.

Builds on

32 of 53 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  3. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  4. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  5. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  6. 2018Attention-based Deep Multiple Instance Learningcited by 1,077
  7. 2016Temporal Ensembling for Semi-Supervised Learningcited by 1,054
  8. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 703
  9. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 473
  10. 2019Implicit Regularization in Deep Matrix Factorizationcited by 402
  11. 2020Combating noisy labels by agreement: A joint training method with co-regularizationcited by 289
  12. 2013Direct and Indirect Effectscited by 276
20 more
  1. 2020Counterfactual VQA: A Cause-Effect Look at Language Biascited by 266
  2. 2020Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effectcited by 256
  3. 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
  4. 2020Representation Learning via Invariant Causal Mechanismscited by 197
  5. 2020Whole Slide Images based Cancer Survival Prediction using Attention Guided Deep Multiple Instance Learning Networkscited by 178
  6. 2017Interpretable Convolutional Neural Networkscited by 161
  7. 2015Patch-based Convolutional Neural Network for Whole Slide Tissue Image Classificationcited by 159
  8. 2016Revisiting Multiple Instance Neural Networkscited by 119
  9. 2021Distilling Causal Effect of Data in Class-Incremental Learningcited by 117
  10. 2020Interventional Few-Shot Learningcited by 112
  11. 2020Visual Commonsense R-CNNcited by 106
  12. 2021Counterfactual Zero-Shot and Open-Set Visual Recognitioncited by 95
  13. 2019Two Causal Principles for Improving Visual Dialogcited by 75
  14. 2019CAMEL: A Weakly Supervised Learning Framework for Histopathology Image Segmentationcited by 73
  15. 2019C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detectioncited by 71
  16. 2018Neural Image Compression for Gigapixel Histopathology Image Analysiscited by 65
  17. 20193C-Net: Category Count and Center Loss for Weakly-Supervised Action Localizationcited by 52
  18. 2019Signet Ring Cell Detection With a Semi-supervised Learning Frameworkcited by 27
  19. 2019Multi-Instance Multi-Scale CNN for Medical Image Classificationcited by 13
  20. 2017Label Stability in Multiple Instance Learningcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 4; 2025: 4; 2026: 1.

13 papers.

Sort:

  1. 2023Interventional Bag Multi-Instance Learning On Whole-Slide Pathological Imagescited by 91
  2. 2024Transcriptomics-guided Slide Representation Learning in Computational Pathologycited by 50
  3. 2023The Rise of AI Language Pathologists: Exploring Two-level Prompt Learning for Few-shot Weakly-supervised Whole Slide Image Classificationcited by 41
  4. 2024Pathology-knowledge Enhanced Multi-instance Prompt Learning for Few-shot Whole Slide Image Classificationcited by 13
  5. 2022Interpreting Vulnerabilities of Multi-Instance Learning to Adversarial Perturbationscited by 3
  6. 2023Data-driven Knowledge Fusion for Deep Multi-instance Learningcited by 3
  7. 2025Cracking Instance Jigsaw Puzzles: An Alternative to Multiple Instance Learning for Whole Slide Image Analysiscited by 3
  8. 2024Rethinking Attention-Based Multiple Instance Learning for Whole-Slide Pathological Image Classification: An Instance Attribute Viewpointcited by 2
  9. 2024Compact and De-biased Negative Instance Embedding for Multi-Instance Learning on Whole-Slide Image Classificationcited by 0
  10. 2025Weakly Semi-supervised Whole Slide Image Classification by Two-level Cross Consistency Supervisioncited by 0
  11. 2025SimMIL: A Universal Weakly Supervised Pre-Training Framework for Multi-Instance Learning in Whole Slide Pathology Imagescited by 0
  12. 2025DLiPath: A Benchmark for the Comprehensive Assessment of Donor Liver Based on Histopathological Image Datasetcited by 0
1 more
  1. 2026Diffusion Attention Expert Model for Predicting and Semi-automatic Localizing STAS in Lung Cancer Histopathological Imagescited by 0