Semi-Supervised Learning with Generative Adversarial Networks

Abstract

We extend Generative Adversarial Networks (GANs) to the semi-supervised context by forcing the discriminator network to output class labels. We train a generative model G and a discriminator D on a dataset with inputs belonging to one of N classes. At training time, D is made to predict which of N+1 classes the input belongs to, where an extra class is added to correspond to the outputs of G. We show that this method can be used to create a more data-efficient classifier and that it allows for generating higher quality samples than a regular GAN.

Appearing in the Data Efficient Machine Learning workshop at ICML 2016

References in corpus (2)

• Conditional Generative Adversarial Nets
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks

Cited by in corpus (122)

• ICLabel: An automated electroencephalographic independent component classifier, dataset, and website
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Temporal Ensembling for Semi-Supervised Learning
• NIPS 2016 Tutorial: Generative Adversarial Networks
• Recent advances and clinical applications of deep learning in medical image analysis
• A Survey on Deep Semi-supervised Learning
• MixMatch: A Holistic Approach to Semi-Supervised Learning
• Sharpness-aware Low dose CT denoising using conditional generative adversarial network
• Generating 3D structures from a 2D slice with GAN-based dimensionality expansion
• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• An Overview of Deep Semi-Supervised Learning
• An empirical study on evaluation metrics of generative adversarial networks
• Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro
• Realistic Evaluation of Deep Semi-Supervised Learning Algorithms
• Learning Adversarially Fair and Transferable Representations
• Representation Learning by Rotating Your Faces
• An Introduction to Image Synthesis with Generative Adversarial Nets
• High-Fidelity Image Generation With Fewer Labels
• 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
• A review of Generative Adversarial Networks for Electronic Health Records: applications, evaluation measures and data sources
• Unsupervised Image-to-Image Translation with Generative Adversarial Networks
• Towards Utilizing Unlabeled Data in Federated Learning: A Survey and Prospective
• Semi-supervised Feature Learning For Improving Writer Identification
• Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Visuomotor Policies
• Multi-View Image Generation from a Single-View
• Two Decades of Bengali Handwritten Digit Recognition: A Survey
• Fully Automatic Electrocardiogram Classification System based on Generative Adversarial Network with Auxiliary Classifier
• Unsupervised and semi-supervised learning with Categorical Generative Adversarial Networks assisted by Wasserstein distance for dermoscopy image Classification
• AnoSeg: Anomaly Segmentation Network Using Self-Supervised Learning
• Adversarial Generation of Training Examples: Applications to Moving Vehicle License Plate Recognition
• Adversarial Training in Affective Computing and Sentiment Analysis: Recent Advances and Perspectives
• Quantum semi-supervised generative adversarial network for enhanced data classification
• Novelty Detection with GAN
• Generative Adversarial Networks (GANs): An Overview of Theoretical Model, Evaluation Metrics, and Recent Developments
• Simple yet Effective Way for Improving the Performance of GAN
• Deep Semi-Supervised Anomaly Detection
• ECG-Adv-GAN: Detecting ECG Adversarial Examples with Conditional Generative Adversarial Networks
• Learning Privacy-Preserving Student Networks via Discriminative-Generative Distillation
• Multi-Channel Attention Selection GAN with Cascaded Semantic Guidance for Cross-View Image Translation
• Towards Discriminability and Diversity: Batch Nuclear-norm Maximization under Label Insufficient Situations
• Learning Loss Functions for Semi-supervised Learning via Discriminative Adversarial Networks
• DiscrimNet: Semi-Supervised Action Recognition from Videos using Generative Adversarial Networks
• FCC-GAN: A Fully Connected and Convolutional Net Architecture for GANs
• SemiCurv: Semi-Supervised Curvilinear Structure Segmentation
• Data Augmentation and Classification of Sea-Land Clutter for Over-the-Horizon Radar Using AC-VAEGAN
• Dual Student: Breaking the Limits of the Teacher in Semi-supervised Learning
• AC-BLSTM: Asymmetric Convolutional Bidirectional LSTM Networks for Text Classification
• Classification of lung cancer subtypes on CT images with synthetic pathological priors
• Semi-supervised Text Regression with Conditional Generative Adversarial Networks
• A Review of Semi Supervised Learning Theories and Recent Advances
• Synthesizing lesions using contextual GANs improves breast cancer classification on mammograms
• A Comparative Review of Recent Few-Shot Object Detection Algorithms
• Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation
• Adversarial symmetric GANs: bridging adversarial samples and adversarial networks
• Rethinking Assumptions in Deep Anomaly Detection
• Score-based Generative Modeling in Latent Space
• On the Generalization Effects of Linear Transformations in Data Augmentation
• Bangla Handwritten Digit Recognition and Generation
• Self-Ensembling Contrastive Learning for Semi-Supervised Medical Image Segmentation
• One-Class Adversarial Nets for Fraud Detection
• cGANs with Multi-Hinge Loss
• Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization
• Cycle In Cycle Generative Adversarial Networks for Keypoint-Guided Image Generation
• EC-GAN: Low-Sample Classification using Semi-Supervised Algorithms and GANs
• Triple Generative Adversarial Networks
• Discriminative Hamiltonian Variational Autoencoder for Accurate Tumor Segmentation in Data-Scarce Regimes
• Conditional GANs with Auxiliary Discriminative Classifier
• Boosting Noise Robustness of Acoustic Model via Deep Adversarial Training
• ESAD: End-to-end Deep Semi-supervised Anomaly Detection
• Self-supervised Semi-supervised Learning for Data Labeling and Quality Evaluation
• DADA: Deep Adversarial Data Augmentation for Extremely Low Data Regime Classification
• cGANs with Conditional Convolution Layer
• Data-Free Learning of Student Networks
• CDCGen: Cross-Domain Conditional Generation via Normalizing Flows and Adversarial Training
• Adversarial Self-Supervised Learning for Semi-Supervised 3D Action Recognition
• Zero-Shot Learning with Knowledge Enhanced Visual Semantic Embeddings
• Tensorizing GAN with High-Order Pooling for Alzheimer's Disease Assessment
• Negative sampling in semi-supervised learning
• Adversarial-based neural networks for affect estimations in the wild
• Snore-GANs: Improving Automatic Snore Sound Classification with Synthesized Data
• DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network
• ByteSGAN: A Semi-supervised Generative Adversarial Network for Encrypted Traffic Classification of SDN Edge Gateway in Green Communication Network
• AutoGAN: Robust Classifier Against Adversarial Attacks
• Joint Learning of Generative Translator and Classifier for Visually Similar Classes
• An interpretable semi-supervised classifier using two different strategies for amended self-labeling
• Modeling EEG data distribution with a Wasserstein Generative Adversarial Network to predict RSVP Events
• Are You A Risk Taker? Adversarial Learning of Asymmetric Cross-Domain Alignment for Risk Tolerance Prediction
• Semi-supervised Learning for Few-shot Image-to-Image Translation
• Adversarial Framework for Unsupervised Learning of Motion Dynamics in Videos
• Emotion Generation and Recognition: A StarGAN Approach
• ScarceGAN: Discriminative Classification Framework for Rare Class Identification for Longitudinal Data with Weak Prior
• Co-supervised learning paradigm with conditional generative adversarial networks for sample-efficient classification
• Generative-Discriminative Complementary Learning
• Generative Adversarial Networks and Adversarial Autoencoders: Tutorial and Survey
• On the link between generative semi-supervised learning and generative open-set recognition
• Likelihood Contribution based Multi-scale Architecture for Generative Flows
• Generative Design of Hardware-aware DNNs
• MU-GAN: Facial Attribute Editing based on Multi-attention Mechanism
• Reliable Semi-Supervised Learning when Labels are Missing at Random
• Typeface Completion with Generative Adversarial Networks
• Read classification using semi-supervised deep learning
• Label Prediction Framework for Semi-Supervised Cross-Modal Retrieval
• TL-SDD: A Transfer Learning-Based Method for Surface Defect Detection with Few Samples
• Hard-Aware Fashion Attribute Classification
• Matching Embeddings for Domain Adaptation
• Geodesics in fibered latent spaces: A geometric approach to learning correspondences between conditions
• An Assessment of GANs for Identity-related Applications
• Semi-Supervised Adversarial Discriminative Domain Adaptation
• Weakly-supervised Generative Adversarial Networks for medical image classification
• Multiple Organ Failure Prediction with Classifier-Guided Generative Adversarial Imputation Networks
• Student's t-Generative Adversarial Networks
• Semi-supervised learning objectives as log-likelihoods in a generative model of data curation
• Data augmentation using generative networks to identify dementia
• A Semi-Supervised Approach for Automatic Crystal Structure Classification
• TRIP: Refining Image-to-Image Translation via Rival Preferences
• Fuzzy Generative Adversarial Networks
• Learning Improved Representations by Transferring Incomplete Evidence Across Heterogeneous Tasks
• Conditional Neural Architecture Search
• Tackling the Problem of Limited Data and Annotations in Semantic Segmentation
• End-To-End Graph-based Deep Semi-Supervised Learning
• Generative Adversarial Nets for Multiple Text Corpora
• Investigating Semi-Supervised Learning Algorithms in Text Datasets