Anomaly Detection using One-Class Neural Networks

Abstract

We propose a one-class neural network (OC-NN) model to detect anomalies in complex data sets. OC-NN combines the ability of deep networks to extract a progressively rich representation of data with the one-class objective of creating a tight envelope around normal data. The OC-NN approach breaks new ground for the following crucial reason: data representation in the hidden layer is driven by the OC-NN objective and is thus customized for anomaly detection. This is a departure from other approaches which use a hybrid approach of learning deep features using an autoencoder and then feeding the features into a separate anomaly detection method like one-class SVM (OC-SVM). The hybrid OC-SVM approach is sub-optimal because it is unable to influence representational learning in the hidden layers. A comprehensive set of experiments demonstrate that on complex data sets (like CIFAR and GTSRB), OC-NN performs on par with state-of-the-art methods and outperformed conventional shallow methods in some scenarios.

References in corpus (4)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• ProtVec: A Continuous Distributed Representation of Biological Sequences
• Robust PCA via Outlier Pursuit

Cited by in corpus (54)

• Deep Learning for Anomaly Detection: A Survey
• Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddings
• Artificial Intelligence based Anomaly Detection of Energy Consumption in Buildings: A Review, Current Trends and New Perspectives
• Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection
• Student-Teacher Feature Pyramid Matching for Anomaly Detection
• Explainable Deep One-Class Classification
• Learning and Evaluating Representations for Deep One-class Classification
• MOCCA: Multi-Layer One-Class ClassificAtion for Anomaly Detection
• On Neural Network Identification for Low-Speed Ship Maneuvering Model
• SSD: A Unified Framework for Self-Supervised Outlier Detection
• Explainable Deep Few-shot Anomaly Detection with Deviation Networks
• A One-Class Classification Decision Tree Based on Kernel Density Estimation
• DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection
• Deep Weakly-supervised Anomaly Detection
• Deep Semi-Supervised Anomaly Detection
• Visual Anomaly Detection in Event Sequence Data
• Towards Visually Explaining Variational Autoencoders
• Reconstruction Student with Attention for Student-Teacher Pyramid Matching
• Deep learning models for predictive maintenance: a survey, comparison, challenges and prospect
• VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localization
• Multi-Modal Anomaly Detection for Unstructured and Uncertain Environments
• Improving Botnet Detection with Recurrent Neural Network and Transfer Learning
• GODS: Generalized One-class Discriminative Subspaces for Anomaly Detection
• Who is in Control? Practical Physical Layer Attack and Defense for mmWave based Sensing in Autonomous Vehicles
• Better the Devil you Know: An Analysis of Evasion Attacks using Out-of-Distribution Adversarial Examples
• Towards Self-Regulating AI: Challenges and Opportunities of AI Model Governance in Financial Services
• Do We Really Need to Learn Representations from In-domain Data for Outlier Detection?
• Explainable Failure Predictions with RNN Classifiers based on Time Series Data
• Semi-supervised Anomaly Detection on Attributed Graphs
• Quick survey of graph-based fraud detection methods
• Out-of-Distribution Dynamics Detection: RL-Relevant Benchmarks and Results
• Self-Supervised Representation Learning for Visual Anomaly Detection
• BEHAVE: Behavior-Aware, Intelligent and Fair Resource Management for Heterogeneous Edge-IoT Systems
• Peek Inside the Closed World: Evaluating Autoencoder-Based Detection of DDoS to Cloud
• Defending Water Treatment Networks: Exploiting Spatio-temporal Effects for Cyber Attack Detection
• Cyber Intrusion Detection by Using Deep Neural Networks with Attack-sharing Loss
• Importance Weighted Adversarial Discriminative Transfer for Anomaly Detection
• AMAD: Adversarial Multiscale Anomaly Detection on High-Dimensional and Time-Evolving Categorical Data
• Sequential Adversarial Anomaly Detection for One-Class Event Data
• $(1+\epsilon )$-class Classification: an Anomaly Detection Method for Highly Imbalanced or Incomplete Data Sets
• DDR-ID: Dual Deep Reconstruction Networks Based Image Decomposition for Anomaly Detection
• Hack The Box: Fooling Deep Learning Abstraction-Based Monitors
• Anomaly Detection Based on Selection and Weighting in Latent Space
• IPOF: An Extremely and Excitingly Simple Outlier Detection Booster via Infinite Propagation
• Understanding the Role of Self-Supervised Learning in Out-of-Distribution Detection Task
• Cross Domain Image Matching in Presence of Outliers
• Open-World Class Discovery with Kernel Networks
• Active Authentication using an Autoencoder regularized CNN-based One-Class Classifier
• ${\ell }_p$-Norm Multiple Kernel One-Class Fisher Null-Space
• New Perspective on Progressive GANs Distillation for One-class Novelty Detection
• Hierarchical Models: Intrinsic Separability in High Dimensions
• GIM: Gaussian Isolation Machines
• Detecting Anomalies Through Contrast in Heterogeneous Data
• Generalized One-Class Learning Using Pairs of Complementary Classifiers