The DOI matches its Crossref record; 221 of the 223 references it cites were found in a registry or an index.
Multiple instance learning: A survey of problem characteristics and applications
Open accessOverview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“With over 30 billion sequences in total, this represents one of the largest multiple instance learning experiments ever conducted (Carbonneau et al., 2018).”
From Hopfield Networks is All You Need (2020) · cites this paper as Carbonneau et al., 2018
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380
SMOTE synthetic minority over-sampling technique
Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002) · Journal of Artificial Intelligence Research · doi:10.1613/jair.953 · Cited by 32,865 · cited 2×
“There are external methods like SMOTE [193] and RUSBoost [194] that necessitate accurate labels to perform over or under sampling.”
MethodGreedy function approximation: A gradient boosting machine
Friedman, J. H. (2001) · The Annals of Statistics · doi:10.1214/aos/1013203451 · Cited by 30,731 · cited 1×
“The method is essentially the same as gradient boosting [143] except that the loss function is based on bag classification error.”
MethodThe pascal visual object classes (VOC) challenge
Everingham, M., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2010) · International Journal of Computer Vision · doi:10.1007/s11263-009-0275-4 · Cited by 19,599
The WEKA data mining software: an update
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., & Witten, I. H. (2009) · ACM SIGKDD Explorations Newsletter · doi:10.1145/1656274.1656278 · Cited by 17,750
Representation Learning: A Review and New Perspectives
Bengio, Y., Courville, A., & Vincent, P. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.50 · Cited by 13,001 · cited 1×
“Recently, methods for representation learning [160] have gained in popularity because they usually yield a high level of accuracy.”
MethodStatistical Comparisons of Classifiers over Multiple Data Sets
Demšar, J. (2006) · doi:10.5555/1248547.1248548 · Cited by 12,051
Cross validatory choice and assessment of statistical predictions
Stone, M. (1974) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1974.tb00994.x · Cited by 10,881 · cited 1×
“The experiments were conducted using a nested cross-fold validation protocol [149].”
MethodBeyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
Lazebnik, S., Schmid, C., & Ponce, J. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR'06) · doi:10.1109/cvpr.2006.68 · Cited by 7,916 · cited 1×
“In BoW models for computer vision, this can be achieved by dividing the images [102, 103] or videos [80] into different spatial and/or temporal zones.”
The Earth Mover's Distance as a Metric for Image Retrieval
Rubner, Y., Tomasi, C., & Guibas, L. (2000) · International Journal of Computer Vision · doi:10.1023/a:1026543900054 · Cited by 5,547
Research works printed but not linked: 1
- Babenko, 2011, Robust object tracking with online multiple instance learning, IEEE Trans. Pattern Anal. Mach. Intell., 33, 1619, 10.1109/TPAMI.2010.226 · DOI · Paper page
Cited by
Sorted by citations
Items 101–200 of 734
Disambiguation Enabled Linear Discriminant Analysis for Partial Label Dimensionality Reduction
Wu, J., & Zhang, M.-L. (2019) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3292500.3330901 · Cited by 34
“a set of instances) instead of individual instances [2, 5, 25].”
BackgroundLópez-Pérez, M., Schmidt, A., Wu, Y., Molina, R., & Katsaggelos, A. K. (2022) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2022.106783 · Cited by 35
A comprehensive survey of intestine histopathological image analysis using machine vision approaches
Jing, Y., Li, C., Du, T., Jiang, T., Sun, H., Yang, J., Shi, L., Gao, M., Grzegorzek, M., & Li, X. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107388 · Cited by 40
Cai, L., Huang, S., Zhang, Y., Lu, J., & Zhang, Y. (2025) · Medical Image Analysis · doi:10.1016/j.media.2025.103631 · Cited by 33
Yacob, F., Siarov, J., Villiamsson, K., Suvilehto, J. T., Sjöblom, L., Kjellberg, M., & Neittaanmäki, N. (2023) · Scientific Reports · doi:10.1038/s41598-023-33863-z · Cited by 33
“Adding these components into MIL approaches has proven to provide better performance 22,23 .”
BackgroundMethodInfluentialEarly Expression Detection via Online Multi-Instance Learning With Nonlinear Extension
Xie, L., Tao, D., & Wei, H. (2018) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2018.2869891 · Cited by 35
“correctly classify unseen instances or bags [26].”
Tavolara, T. E., Niazi, M. K. K., Ginese, M. L., Piedra-Mora, C., Gatti, D. M., Beamer, G., & Gürcan, M. N. (2020) · EBioMedicine · doi:10.1016/j.ebiom.2020.103094 · Cited by 33
“Classification by MIL is performed at the bag level and not the single instance level like in supervised learning [52].”
BackgroundBlokker, M., Hamer, P. C. D. W., Wesseling, P., Groot, M. L., & Veta, M. (2022) · Scientific Reports · doi:10.1038/s41598-022-15423-z · Cited by 33
DSCA: A dual-stream network with cross-attention on whole-slide image pyramids for cancer prognosis
Liu, P., Fu, B., Ye, F., Yang, R., & Ji, L. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120280 · Cited by 32
“Multiple Instance Learning (MIL) is a weakly-supervised learning framework (Dietterich et al., 1997; Carbonneau et al., 2018).”
BackgroundSauter, D., Lodde, G. C., Nensa, F., Schadendorf, D., Livingstone, E., & Kukuk, M. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107083 · Cited by 37
Distilling Privileged Knowledge for Anomalous Event Detection From Weakly Labeled Videos
Liu, T., Lam, K., & Kong, J. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3263966 · Cited by 32
Machine Learning for the Semantic Web: Lessons learnt and next research directions
d’Amato, C. (2020) · Semantic Web · doi:10.3233/sw-200388 · Cited by 32
“Multiple-instance learning (MIL) [56] is also a setting that would need investigation.”
BackgroundInfluentialPartial Label Dimensionality Reduction via Confidence-Based Dependence Maximization
Bao, W.-X., Hang, J.-Y., & Zhang, M.-L. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467313 · Cited by 32
“In terms of the formulation of the learning problem, partial label learning is related to several well-established weakly supervised learning frameworks such as multi-label learning [44, 50, 51], semi-supervised learning [55, 58] and multi-instance learning [2, 5].”
BackgroundWeakly-Supervised Learning for Fine-Grained Emotion Recognition Using Physiological Signals
Zhang, T., Ali, A. E., Wang, C., Hanjalić, A., & Cesar, P. S. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2022.3158234 · Cited by 35
Shen, B., Feng, F., Li, K., Lin, P., Ma, L., & Li, H. (2022) · Briefings in Bioinformatics · doi:10.1093/bib/bbac605 · Cited by 31
Barbera, D. L., Polónia, A., Roitero, K., Conde‐Sousa, E., & Mea, V. D. (2020) · Journal of Imaging · doi:10.3390/jimaging6090082 · Cited by 32
“A technique that is used to overcome such limitations is represented by MIL [23,24].”
MethodBollis, E., Maia, H. D. A., Pedrini, H., & Avila, S. (2022) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2022.106839 · Cited by 31
Random forest feature selection for partial label learning
Sun, X., & Chai, J. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126870 · Cited by 30
Shao, Z., Wang, Y., Chen, Y., Bian, H., Liu, S., Wang, H., & Zhang, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01965 · Cited by 31
“In this paper, we describe this label ambiguity [9] as a semantical unalignment between patches and WSI-level label.”
MethodOn the Usefulness of Pre-Processing Step in Melanoma Detection Using Multiple Instance Learning
Vocaturo, E., Zumpano, E., & Veltri, P. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-27629-4_34 · Cited by 30
Deep Weakly Supervised Domain Adaptation for Pain Localization in Videos
Praveen, R. G., Granger, É., & Cardinal, P. (2020) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg47880.2020.00139 · Cited by 30
“Multiple Instance Learning (MIL) is one of the widely used approaches for inexact supervision [6].”
BackgroundShort-Term Strong Wind Risk Prediction for High-Speed Railway
Liu, H., Liu, C., He, S., & Chen, J. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3058608 · Cited by 32
Deep multi‐instance transfer learning for pneumothorax classification in chest X‐ray images
Tian, Y., Wang, J., Yang, W., Wang, J., & Qian, D. (2021) · Medical Physics · doi:10.1002/mp.15328 · Cited by 32
Border, S. P., & Sarder, P. (2022) · Frontiers in Physiology · doi:10.3389/fphys.2021.821217 · Cited by 31
“…Density (EM-DD) and Axis-Parallel Rectangles, have demonstrated significant performance in defining bag-level distributions of data given feature values for weakly supervised tasks (Dietterich et al., 1997; Zhang and Goldman, 2001; Foulds and Frank, 2010; Carbonneau et al., 2018).”
BackgroundZhang, X., Li, C., Chen, W., Zheng, J., & Li, F. (2025) · Scientific Reports · doi:10.1038/s41598-025-90117-w · Cited by 37
“Multiple instance learning (MIL) is a variant of supervised learning designed to solve the classification problem when individual instance labels are unavailable or difficult to obtain 24 .”
Discriminative Multi-View Dynamic Image Fusion for Cross-View 3-D Action Recognition
Wang, Y., Xiao, Y., Lu, J., Tan, B., Cao, Z., Zhang, Z., & Zhou, J. T. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3070179 · Cited by 29
“To address this, we propose to formulate MVDI fusion task as a multi-instance learning (MIL) problem [18], [19].”
BackgroundMethodMachine Learning for Fraud Detection in E-Commerce: A Research Agenda
Tax, N., de Vries, K. J., de Jong, M., Dosoula, N., van den Akker, B., Smith, J., Thuong, O., & Bernardi, L. (2021) · Communications in computer and information science · doi:10.1007/978-3-030-87839-9_2 · Cited by 39
BiLabel-Specific Features for Multi-Label Classification
Zhang, M.-L., Fang, J.-P., & Wang, Y. (2021) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3458283 · Cited by 29
“In multi-instance classification [1, 7, 25], a single class label is assigned to a bag of instances while for multi-label classification a number of class labels are assigned to a single instance.”
BackgroundWaqas, M., Tahir, M. A., Author, M. D., Al-Maadeed, S., Bouridane, A., & Wu, J. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-09417-3 · Cited by 28
Chikontwe, P., Kim, M., Jeong, J., Sung, H. J., Go, H., Nam, S. J., & Park, S. H. (2024) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2024.3446716 · Cited by 31
“In contrast to standard MIL aggregators [5], [21], [34] such as max-or mean-pooling, FR-MIL learns to create a single bag representation via a self-attention mechanism [8] agnostic to the order of instances.”
Otesteanu, C. F., Ugrinic, M., Holzner, G., Chang, Y., Fassnacht, C., Guenova, E., Stavrakis, S., deMello, A. J., & Claassen, M. (2021) · Cell Reports Methods · doi:10.1016/j.crmeth.2021.100094 · Cited by 28
“, 2008) in the cell images, given that it does not use any prior assumptions regarding the data (Carbonneau et al., 2018).”
BackgroundBayesian Collaborative Learning for Whole-Slide Image Classification
Yu, J.-G., Wu, Z., Ming, Y., Deng, S., Wu, Q., Xiong, Z., Yu, T., Xia, G.-S., Jiang, Q., & Li, Y. (2023) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2023.3241204 · Cited by 29
“Multiple instance learning (MIL) [6], [7], [12], [13], [17], [18], [19], [20], [21] is widely treated as a promising way to conquer these challenges.”
BackgroundInfluentialXiong, D., Zhang, Z., Wang, T., & Wang, X. (2021) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2021.05.038 · Cited by 29
“[22], positive bags with low WRs may result in poor performance of many MIL methods as positive and negative bags become similar.”
BackgroundMethodInfluentialMILL: Channel Attention–based Deep Multiple Instance Learning for Landslide Recognition
Tang, X., Liu, M., Zhong, H., Ju, Y., Li, W., & Xu, Q. (2021) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3454009 · Cited by 29
“MIL is a form of weakly supervised learning where training instances are arranged in sets, called bags, and a label is provided for the entire bag [Carbonneau et al. 2018].”
BackgroundBenchmarking Jetson Edge Devices with an End-to-End Video-Based Anomaly Detection System
Pham, H. V., Tran, T. G., Le, C. D., Le, A. D., & Vo, H. B. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-53963-3_25 · Cited by 26
“Like Multi-Instance Learning (MIL) [12], RTFM learns from weakly labeled videos to identify whether a snippet is normal or abnormal.”
BackgroundA multimodal dataset for precision oncology in head and neck cancer
Dörrich, M., Balk, M., Heusinger, T., Beyer, S., Mirbagheri, H., Fischer, D. J., Kanso, H., Matek, C., Hartmann, A., Iro, H., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-62386-6 · Cited by 28
“The first split defined ”in distribution” cases as test data, the second split defined ”out of distribution” data as test data, and the third split defined cases with oropharyngeal cancer as test data (see Figure 2D.”
Dangerousness of dysplastic nevi: a Multiple Instance Learning Solution for Early Diagnosis
Vocaturo, E., & Zumpano, E. (2019) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm47256.2019.8983056 · Cited by 26
“For further details and general considerations on the MIL paradigm, we refer the reader to surveys [33], [34].”
BackgroundLiu, H., Xu, W., Shang, Z.-H., Wang, X., Zhou, H.-Y., Ma, K.-W., Zhou, H., Qi, J.-L., Jiang, J., Tan, L.-L., et al. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.858453 · Cited by 26
Waqas, M., Tahir, M. A., & Qureshi, R. (2022) · Applied Intelligence · doi:10.1007/s10489-022-04045-7 · Cited by 26
“The MIL problems are formulated as bag or instance level classification [3].”
BackgroundMethodCT-based COPD identification using multiple instance learning with two-stage attention
Xue, M., Jia, S., Chen, L., Huang, H., Lijuan, Y., & Zhu, W. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107356 · Cited by 27
Audiovisual Dependency Attention for Violence Detection in Videos
Pang, W., Xie, W. C., He, Q., Li, Y., & Yang, J. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3184533 · Cited by 25
“Sultani et al. [1] employed visual features extracted from a 3D-convolutional network [11] and multiple-instance learning (MIL) [12], [13].”
MethodSelf-supervised adversarial adaptation network for breast cancer detection
Torabi, M., Rasouli, A., Wu, Q. M. J., Cao, W., & Pourpanah, F. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.108489 · Cited by 26
A Multiple Instance Learning Algorithm for Color Images Classification
Astorino, A., Fuduli, A., Gaudioso, M., & Vocaturo, E. (2018) · International Database Engineering and Applications Symposium (IDEAS) · doi:10.1145/3216122.3216144 · Cited by 25
“In this paper, we face image classification by means of a Multiple Instance Learning approach [1, 8] based on a Lagrangian relaxation technique [5].”
BackgroundMethodDiagnosing Dysarthria with Long Short-Term Memory Networks
Mayle, A., Mou, Z., Bunescu, R., Mirshekarian, S., Xu, L., & Liu, C. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-2903 · Cited by 25
Multi-View And Multi-Modal Event Detection Utilizing Transformer-Based Multi-Sensor Fusion
Yasuda, M., Ohishi, Y., Saito, S., & Harado, N. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746006 · Cited by 25
Yang, Y., Li, W., Guo, Y., Zeng, N., Wang, S., Chen, Z., Liu, Y., Chen, H., Duan, W., Li, X., et al. (2022) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2022366 · Cited by 25
Attention-guided MIL weakly supervised visual anomaly detection
Wang, L., Wang, X., Liu, F., Li, M., Hao, X., & Zhao, N. (2023) · Measurement · doi:10.1016/j.measurement.2023.112500 · Cited by 25
(2020) · doi:10.1145/3351095.3375784 · Cited by 29
“Lastly, deep features , generated by a recursive neural network that condenses information of the individual-level features into a one-dimensional encoding—a technique akin to the AI sub-field of multiple instance learning (MIL) [5].”
MethodIterative multiple instance learning for weakly annotated whole slide image classification
Zhou, Y., Che, S., Lü, F., Liu, S., Yan, Z., Wei, J., Li, Y., Ding, X., & Lu, Y. (2023) · Physics in Medicine and Biology · doi:10.1088/1361-6560/acde3f · Cited by 24
Lutz, B., Pröllochs, N., & Neumann, D. (2019) · Hawaii International Conference on System Sciences (HICSS) · doi:10.24251/hicss.2019.137 · Cited by 24
Gonthier, N., Ladjal, S., & Gousseau, Y. (2021) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2021.103299 · Cited by 29
“From a recent survey [50] on multiple Instance Learning it appears that the most efficient algorithm for an instance level classification seems to be a clever variation of bagging and multiple classifiers to deal with multi-modal distributions [51].”
BackgroundDeep multiple instance selection
Li, X., Zhan, D., Yang, J., & Shi, Y. (2021) · Science China Information Sciences · doi:10.1007/s11432-020-3117-3 · Cited by 24
Teramoto, A., Kiriyama, Y., Tsukamoto, T., Sakurai, E., Michiba, A., Imaizumi, K., Saito, K., & Fujita, H. (2021) · Scientific Reports · doi:10.1038/s41598-021-99246-4 · Cited by 24
Assignments as Influential Factor to Improve the Prediction of Student Performance in Online Courses
Esteban, A., Romero, C., & Zafra, A. (2021) · Applied Sciences · doi:10.3390/app112110145 · Cited by 24
“[12] to represent complicated objects [14].”
MethodDisambiguation Enabled Linear Discriminant Analysis for Partial Label Dimensionality Reduction
Zhang, M.-L., Wu, J., & Bao, W.-X. (2022) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3494565 · Cited by 24
CNN-ViT Supported Weakly-Supervised Video Segment Level Anomaly Detection
Sharif, M. H., Jiao, L., & Omlin, C. W. P. (2023) · Sensors · doi:10.3390/s23187734 · Cited by 23
“Supervision is provided solely for complete sets, and the isolated label of the snippets contained in the bags is not provided [17].”
BackgroundA survey on intelligent management of alerts and incidents in IT services
Yu, Q., Zhao, N., Li, M., Li, Z., Wang, H., Zhang, W., Sui, K., & Pei, D. (2024) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2024.103842 · Cited by 38
Zhang, M., Cheng, Q., Wei, Z., Xu, J., Wu, S., Xu, N., Zhao, C., Yu, L., & Feng, W. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae420 · Cited by 23
Naso, J. R., Levine, A., Farahani, H., Chirieac, L. R., Đačić, S., Wright, J. L., Lai, C. K., Yang, H.-M., Jones, S. J. M., Bashashati, A., et al. (2021) · Modern Pathology · doi:10.1038/s41379-021-00850-6 · Cited by 23
Liu, P., Ji, L., Ye, F., & Fu, B. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107433 · Cited by 25
Deep anomaly detection on set data: Survey and comparison
Mašková, M., Zorek, M., Pevný, T., & Šmídl, V. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110381 · Cited by 25
Meta-Black-Box optimization for evolutionary algorithms: Review and perspective
Yang, X., Wang, R., Li, K., & Ishibuchi, H. (2025) · Swarm and Evolutionary Computation · doi:10.1016/j.swevo.2024.101838 · Cited by 27
Aesthetical Issues of Leonardo Da Vinci’s and Pablo Picasso’s Paintings with Stochastic Evaluation
Sargentis, G.-F., Dimitriadis, P., & Koutsoyiannis, D. (2020) · Heritage · doi:10.3390/heritage3020017 · Cited by 22
Weakly supervised learning and interpretability for endometrial whole slide image diagnosis
Mohammadi, M., Cooper, J. R., Arandjelović, O., Fell, C., MORRISON, D. L., Syed, S., Konanahalli, P., Bell, S., Bryson, G., Harrison, D. J., et al. (2022) · Experimental Biology and Medicine · doi:10.1177/15353702221126560 · Cited by 21
Chen, Y., Shao, Z., Bian, H., Fang, Z., Wang, Y., Cai, Y., Wang, H., Liu, G., Li, X., & Zhang, Y. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3285275 · Cited by 22
“For bag classification, embedding-level MIL methods usually perform better than instance-level MIL methods [20].”
MethodSupervised Classification: Quite a Brief Overview
Loog, M. (2018) · Elsevier eBooks · doi:10.1016/b978-0-12-811788-0.00005-6 · Cited by 28
No pixel-level annotations needed
van der Laak, J. A. W. M., Ciompi, F., & Litjens, G. J. S. (2019) · Nature Biomedical Engineering · doi:10.1038/s41551-019-0472-6 · Cited by 22
Cheng, H., Huang, S., Cai, L., Xu, Y., Wang, R., & Zhang, Y. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3528625 · Cited by 24
“However, the instance-level pseudo-labels derived from bag-level labels usually contain a lot of noise [32], which impairs the final classification performance.”
Multiple instance learning based deep CNN for image memorability prediction
Basavaraju, S., & Sur, A. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-019-08202-y · Cited by 21
“MIL is a weakly supervised learning framework where ground truth label is provided for the entire bag and no label information is provided for the instances within each bag [7].”
BackgroundMulticlass Multiple-Instance Learning for Predicting Precursors to Aviation Safety Events
Bleu-Laine, M.-H., Puranik, T. G., Mavris, D. N., & Matthews, B. L. (2021) · Journal of Aerospace Information Systems · doi:10.2514/1.i010971 · Cited by 21
Moranguinho, J., Pereira, T., Ramos, B., Morgado, J., Costa, J. L., & Oliveira, H. P. (2021) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc46164.2021.9631000 · Cited by 21
“Multiple Instance Learning (MIL) is considered a supervised/weakly supervised Machine Learning technique where data gets regarded as an amalgamation of smaller components [7], [8].”
MethodXia, W., Li, D., He, W., Pickhardt, P. J., Jian, J., Zhang, R., Zhang, J., Song, R., Tong, T., Yang, X., et al. (2024) · Radiology Artificial Intelligence · doi:10.1148/ryai.230152 · Cited by 21
“Inspired by weakly supervised learning (14,21,22), a weak supervision component was designed for matching the LN-level metastatic probability to the patient-level label, as follows: (a) the multiple-instance learning strategy implemented by the global maximum pooling layer: the LN with the highest…”
MethodProDiv: Prototype-driven consistent pseudo-bag division for whole-slide image classification
Yang, R., Liu, P., & Ji, L. (2024) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2024.108161 · Cited by 22
“Among them, the weakly-supervised framework, based on embedding-level multiple instance learning (MIL) [4,12], has been extensively studied in recent years.”
BackgroundShin, J., Kaneko, Y., Miah, A. S. M., Hassan, N., & Nishimura, S. (2024) · IEEE Access · doi:10.1109/access.2024.3395329 · Cited by 21
“Super-vision is solely provided for complete sets, and the individual labels of the snippets within the bags are not given [17].”
BackgroundCMAL: Cost-Effective Multi-Label Active Learning by Querying Subexamples
Yu, G., Chen, X., Domeniconi, C., Wang, J., Li, Z., Zhang, Z., & Zhang, X. (2020) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2020.3003899 · Cited by 21
“From the MIL perspective [45], [46], [51], the complex example can also be modeled as a bag and its subexamples can be viewed as instances of that bag.”
BackgroundHooven, T. A., Lin, Y. C., & Salleb-Aouissi, A. (2020) · ACM Conference on Health, Inference, and Learning (CHIL) · doi:10.1145/3368555.3384466 · Cited by 20
“MIL is a form of weakly supervised learning [7, 10], where the training instances are arranged in sets, called bags, and the label is provided for the entire bag.”
BackgroundHAMIL: Hierarchical aggregation-based multi-instance learning for microscopy image classification
Yang, Y., Tu, Y., Lei, H., & Long, W. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.109245 · Cited by 22
Yi, S., Fan, Z., & Wu, D. (2022) · Image and Vision Computing · doi:10.1016/j.imavis.2022.104397 · Cited by 20
Isoform function prediction by Gene Ontology embedding
Qiu, S., Yu, G., Lü, X., Domeniconi, C., & Guo, M. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac576 · Cited by 20
Meta Multi-Instance Multi-Label learning by heterogeneous network fusion
Qiu, S., Wang, M., Yang, Y., Yu, G., Wang, J., Yan, Z., Domeniconi, C., & Guo, M. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.02.010 · Cited by 20
Sajjad, U., Rezapour, M., Su, Z., Tozbikian, G. H., Gürcan, M. N., & Niazi, M. K. K. (2023) · Cancers · doi:10.3390/cancers15133428 · Cited by 20
“Unlike fully supervised learning methods, weak labels are assigned to the bag rather than the individual instances [11].”
BackgroundFeyisa, D. W., Ayano, Y. M., Debelee, T. G., & Schwenker, F. (2023) · Sensors · doi:10.3390/s23156781 · Cited by 20
“The model also learns to identify which pixels in the bag contributed to the prediction of the class label [67].”
BackgroundAnaya, J., Sidhom, J.-W., Mahmood, F., & Baras, A. S. (2023) · Nature Biomedical Engineering · doi:10.1038/s41551-023-01120-3 · Cited by 24
Multi-instance partial-label learning: towards exploiting dual inexact supervision
Tang, W., Zhang, W., & Zhang, M.-L. (2024) · Science China Information Sciences · doi:10.1007/s11432-023-3771-6 · Cited by 21
“As illustrated in Figure 1a, in MIL, multiple training instances are arranged in a bag and we only know the binary bag-level label rather than the instance-level labels [1, 4].”
MethodWang, G., Zhang, X., Peng, Z., Zhang, T., Tang, X., Zhou, H., & Jiao, L. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3395751 · Cited by 20
“Following the definition of MIL that positive bags contain at least one positive instance while negative bags only involve negative instances [60], bag-level label in the case of binary classification is thus defined as follows : where y i is the instance-level label but unknown during the training…”
Yang, Z., Zhou, H., Srivastav, S., Shaffer, J. G., Abraham, K. T., Naandam, S. M., & Kakraba, S. (2025) · Data · doi:10.3390/data10010004 · Cited by 22
Classification of Volumetric Images Using Multi-Instance Learning and Extreme Value Theorem
Tennakoon, R., Bortsova, G., Ørting, S. N., Gostar, A. K., Wille, M. M. W., Saghir, Z., Hoseinnezhad, R., de Bruijne, M., & Bab‐Hadiashar, A. (2019) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2019.2936244 · Cited by 20
“Learning sub-optimal decision functions is a known limitation of MIL or weakly supervised learning methods and detailed analysis of this shortcoming can be found in [33], [45].”
BackgroundMethodOstmeyer, J., Lucas, E., Christley, S., Lea, J. S., Monson, N. L., Tiro, J. A., & Cowell, L. G. (2020) · PLoS ONE · doi:10.1371/journal.pone.0229569 · Cited by 20
“Utilizing publicly available TCR deep sequencing data, we applied multiple instance learning (MIL) after converting the TCR amino acid sequences to a biophysicochemical representation using Atchley Factors [23–25].”
BackgroundInfluentialA Novel Attribute-Based Symmetric Multiple Instance Learning for Histopathological Image Analysis
Vu, T., Lai, P., Raich, R., Pham, A. T., Fern, X. Z., & Rao, U. A. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.2987796 · Cited by 20
“Therefore, researchers have recently shown a general interest in other more loose assumptions such as the collective assumption [37], [38].”
BackgroundMultiple instance learning algorithm for medical image classification
Astorino, A., Fuduli, A., Gaudioso, M., & Vocaturo, E. (2019) · Cited by 19
From Weakly Supervised Learning to Biquality Learning: an Introduction
Nodet, P., Lemaire, V., Bondu, A., Cornuéjols, A., & Ouorou, A. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9533353 · Cited by 23
“This is the case for instance, in Multi Instance Learning (MIL) [3]–[6], in which there is one label for each bag of training examples, and each example has an uncertain label.”
MethodSelf-Attention Pooling-Based Long-Term Temporal Network for Action Recognition
Li, H., Huang, J., Zhou, M., Shi, Q., & Fei, Q. (2022) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2022.3145839 · Cited by 19
“For example, multiple instance learning (MIL) methods are presented to extract temporal features across an entire video, or identify key volumes within the frame sequences [20].”
MethodLeiby, J. S., Hao, J., Kang, G. H., Park, J. W., & Kim, D. (2022) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc48229.2022.9871553 · Cited by 18
“Multiple instance learning (MIL) is a weakly-supervised learning framework where instead of assigning a label to each instance, the instances are grouped into collections, bags, and each bag is assigned a label [16].”
BackgroundChemical complexity challenge: Is multi‐instance machine learning a solution?
Zankov, D. V., Madzhidov, T., Varnek, A. A., & Полищук, П. Г. (2023) · Wiley Interdisciplinary Reviews Computational Molecular Science · doi:10.1002/wcms.1698 · Cited by 19
“9,45”
BackgroundFedmm: Federated Multi-Modal Learning with Modality Heterogeneity in Computational Pathology
Peng, Y., Bian, J., & Xu, J. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10448255 · Cited by 25
Wang, M., Tie, C., Wang, J., Zhu, J., Chen, B., Li, Y., Zhang, S., Liu, L., Guo, L., Yang, L., et al. (2024) · American Journal of Otolaryngology · doi:10.1016/j.amjoto.2024.104342 · Cited by 19
Marini, N., Marchesin, S., Wodziński, M., Caputo, A., Podareanu, D., Guevara, B. C., Boytcheva, S., Vatrano, S., Fraggetta, F., Ciompi, F., et al. (2024) · Medical Image Analysis · doi:10.1016/j.media.2024.103303 · Cited by 20
Deep blind image quality assessment based on multiple instance regression
Liang, D., Gao, X., Lu, W. B., & Li, J. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.12.009 · Cited by 19
Multiple Conformer Descriptors for QSAR Modeling
Nikonenko, A., Zankov, D. V., Baskin, I. I., Madzhidov, T., & Полищук, П. Г. (2021) · Molecular Informatics · doi:10.1002/minf.202060030 · Cited by 19
Metadata
- Authors
- Carbonneau, M.-A., Cheplygina, V., Granger, E., & Gagnon, G.
- Affiliations
- École de Technologie Supérieure (CA)
- Erasmus MC (NL)
- Corresponding author
- Carbonneau, Marc-André
- Venue
- Pattern Recognition
- Publisher
- Elsevier BV
- Published
- 2018-05
- Volume
- 77
- Pages
- 329-353
- ISSN
- 0031-3203 (ISSN-L)
- OpenAlex
- W2560886373
- Cited by
- 740
- References
- 223
- Citations by year
- 2026: 75 · 2025: 118 · 2024: 91 · 2023: 124 · 2022: 98 · 2021: 92 · 2020: 84 · 2019: 41 · 2018: 14
- FWCI
- 22.10
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Pattern Recognition (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- TU/e Research Portal (submittedVersion)
- EUR Research Repository (Erasmus University Rotterdam) (submittedVersion)
- Espace ÉTS (ETS) (submittedVersion)
- TU/e Research Portal (submittedVersion, other-oa)
- RePub (Erasmus University Rotterdam) (submittedVersion, other-oa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- multiple instance learning, weakly supervised learning, problem characteristics, data distribution, label ambiguity, computer vision, document classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
