Multiple Instance Learning
233 papers with code • 0 benchmarks • 8 datasets
Multiple Instance Learning is a type of weakly supervised learning algorithm where training data is arranged in bags, where each bag contains a set of instances $X=\{x_1,x_2, \ldots,x_M\}$, and there is one single label $Y$ per bag, $Y\in\{0, 1\}$ in the case of a binary classification problem. It is assumed that individual labels $y_1, y_2,\ldots, y_M$ exist for the instances within a bag, but they are unknown during training. In the standard Multiple Instance assumption, a bag is considered negative if all its instances are negative. On the other hand, a bag is positive, if at least one instance in the bag is positive.
Source: Monte-Carlo Sampling applied to Multiple Instance Learning for Histological Image Classification
Benchmarks
These leaderboards are used to track progress in Multiple Instance Learning
Libraries
Use these libraries to find Multiple Instance Learning models and implementations
UCF-Crime
AVE
SPOT
CamNuvem Dataset
Latest papers
Multi-head Attention-based Deep Multiple Instance Learning
tueimage/mad-mil
•
•
This paper introduces MAD-MIL, a Multi-head Attention-based Deep Multiple Instance Learning model, designed for weakly supervised Whole Slide Images (WSIs) classification in digital pathology.
DinoBloom: A Foundation Model for Generalizable Cell Embeddings in Hematology
In hematology, computational models offer significant potential to improve diagnostic accuracy, streamline workflows, and reduce the tedious work of analyzing single cells in peripheral blood or bone marrow smears.
Hyperbolic Secant representation of the logistic function: Application to probabilistic Multiple Instance Learning for CT intracranial hemorrhage detection
This approach yields the same variational posterior approximations as the original VGPMIL, which is a consequence of the two representations that the Hyperbolic Secant distribution admits.
Counting Network for Learning from Majority Label
shiku-kaito/counting-network-for-learning-from-majority-label
•
•
Existing MIL methods are unsuitable for LML due to aggregating confidences, which may lead to inconsistency between the bag-level label and the label obtained by counting the number of instances for each class.
MambaMIL: Enhancing Long Sequence Modeling with Sequence Reordering in Computational Pathology
Multiple Instance Learning (MIL) has emerged as a dominant paradigm to extract discriminative feature representations within Whole Slide Images (WSIs) in computational pathology.
HistGen: Histopathology Report Generation via Local-Global Feature Encoding and Cross-modal Context Interaction
Histopathology serves as the gold standard in cancer diagnosis, with clinical reports being vital in interpreting and understanding this process, guiding cancer treatment and patient care.
Generalizable Whole Slide Image Classification with Fine-Grained Visual-Semantic Interaction
ls1rius/wsi_five
•
•
It is designed to enhance the model's generalizability by leveraging the interaction between localized visual patterns and fine-grained pathological semantics.
Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology
Unlike existing works that focus on pre-training powerful feature extractor or designing sophisticated instance aggregator, R$^2$T is tailored to re-embed instance features online.
Sparse and Structured Hopfield Networks
deep-spin/sshn
•
•
Modern Hopfield networks have enjoyed recent interest due to their connection to attention in transformers.
Weakly Supervised Object Detection in Chest X-Rays with Differentiable ROI Proposal Networks and Soft ROI Pooling
Weakly supervised object detection (WSup-OD) increases the usefulness and interpretability of image classification algorithms without requiring additional supervision.
