Revisiting Multiple Instance Neural Networks

Abstract

Recently neural networks and multiple instance learning are both attractive topics in Artificial Intelligence related research fields. Deep neural networks have achieved great success in supervised learning problems, and multiple instance learning as a typical weakly-supervised learning method is effective for many applications in computer vision, biometrics, nature language processing, etc. In this paper, we revisit the problem of solving multiple instance learning problems using neural networks. Neural networks are appealing for solving multiple instance learning problem. The multiple instance neural networks perform multiple instance learning in an end-to-end way, which take a bag with various number of instances as input and directly output bag label. All of the parameters in a multiple instance network are able to be optimized via back-propagation. We propose a new multiple instance neural network to learn bag representations, which is different from the existing multiple instance neural networks that focus on estimating instance label. In addition, recent tricks developed in deep learning have been studied in multiple instance networks, we find deep supervision is effective for boosting bag classification accuracy. In the experiments, the proposed multiple instance networks achieve state-of-the-art or competitive performance on several MIL benchmarks. Moreover, it is extremely fast for both testing and training, e.g., it takes only 0.0003 second to predict a bag and a few seconds to train on a MIL datasets on a moderate CPU.

References in corpus (1)

• Relaxed Multiple-Instance SVM with Application to Object Discovery

Cited by in corpus (52)

• Attention-based Deep Multiple Instance Learning
• Evaluate the Malignancy of Pulmonary Nodules Using the 3D Deep Leaky Noisy-or Network
• Monte-Carlo Sampling applied to Multiple Instance Learning for Histological Image Classification
• RMDL: Recalibrated multi-instance deep learning for whole slide gastric image classification
• Sharp Multiple Instance Learning for DeepFake Video Detection
• Multiple instance dense connected convolution neural network for aerial image scene classification
• Weakly Labelled AudioSet Tagging with Attention Neural Networks
• Detecting Parkinsonian Tremor from IMU Data Collected In-The-Wild using Deep Multiple-Instance Learning
• Hopfield Networks is All You Need
• Towards Label-efficient Automatic Diagnosis and Analysis: A Comprehensive Survey of Advanced Deep Learning-based Weakly-supervised, Semi-supervised and Self-supervised Techniques in Histopathological Image Analysis
• Multiple instance learning with graph neural networks
• All Grains, One Scheme (AGOS): Learning Multi-grain Instance Representation for Aerial Scene Classification
• Deep Weakly-Supervised Learning Methods for Classification and Localization in Histology Images: A Survey
• Explainable Deep Few-shot Anomaly Detection with Deviation Networks
• Lung Cancer Diagnosis Using Deep Attention Based on Multiple Instance Learning and Radiomics
• Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning
• DSCA: A Dual-Stream Network with Cross-Attention on Whole-Slide Image Pyramids for Cancer Prognosis
• PS-DeVCEM: Pathology-sensitive deep learning model for video capsule endoscopy based on weakly labeled data
• Attention2Minority: A salient instance inference-based multiple instance learning for classifying small lesions in whole slide images
• Multiple instance learning on deep features for weakly supervised object detection with extreme domain shifts
• Sharing Pain: Using Pain Domain Transfer for Video Recognition of Low Grade Orthopedic Pain in Horses
• An embarrassingly simple approach to neural multiple instance classification
• Agent Aggregator with Mask Denoise Mechanism for Histopathology Whole Slide Image Analysis
• ProtoDiv: Prototype-guided Division of Consistent Pseudo-bags for Whole-slide Image Classification
• Attention-based Multi-instance Neural Network for Medical Diagnosis from Incomplete and Low Quality Data
• Psychophysiological Arousal in Young Children Who Stutter: An Interpretable AI Approach
• Weakly Supervised Clustering by Exploiting Unique Class Count
• Studying The Effect of MIL Pooling Filters on MIL Tasks
• Hard Sample Mining for the Improved Retraining of Automatic Speech Recognition
• AMIL: Adversarial Multi Instance Learning for Human Pose Estimation
• Dual-stream Maximum Self-attention Multi-instance Learning
• Inner-Imaging Networks: Put Lenses into Convolutional Structure
• AMI-Net+: A Novel Multi-Instance Neural Network for Medical Diagnosis from Incomplete and Imbalanced Data
• Multi-Instance Learning by Utilizing Structural Relationship among Instances
• AMINN: Autoencoder-based Multiple Instance Neural Network Improves Outcome Prediction of Multifocal Liver Metastases
• One Size Fits Many: Column Bundle for Multi-X Learning
• Specialized Decision Surface and Disentangled Feature for Weakly-Supervised Polyphonic Sound Event Detection
• Multiple Instance Learning for Heterogeneous Images: Training a CNN for Histopathology
• A bag-to-class divergence approach to multiple-instance learning
• Non-I.I.D. Multi-Instance Learning for Predicting Instance and Bag Labels using Variational Auto-Encoder
• CNN based Multi-Instance Multi-Task Learning for Syndrome Differentiation of Diabetic Patients
• A Global-local Attention Framework for Weakly Labelled Audio Tagging
• LogAvgExp Provides a Principled and Performant Global Pooling Operator
• Deep Learning from Label Proportions for Emphysema Quantification
• Cross-scale Attention Model for Acoustic Event Classification
• Learning Maximally Predictive Prototypes in Multiple Instance Learning
• Sparse Network Inversion for Key Instance Detection in Multiple Instance Learning
• A Real-time Action Representation with Temporal Encoding and Deep Compression
• In Defense of LSTMs for Addressing Multiple Instance Learning Problems
• A deep multiple instance learning approach based on coarse labels for high-resolution land-cover mapping
• Weakly Supervised Object Localization and Detection: A Survey
• COVID-view: Diagnosis of COVID-19 using Chest CT