Meta-learning One-class Classifiers with Eigenvalue Solvers for Supervised Anomaly Detection

Abstract

Neural network-based anomaly detection methods have shown to achieve high performance. However, they require a large amount of training data for each task. We propose a neural network-based meta-learning method for supervised anomaly detection. The proposed method improves the anomaly detection performance on unseen tasks, which contains a few labeled normal and anomalous instances, by meta-training with various datasets. With a meta-learning framework, quick adaptation to each task and its effective backpropagation are important since the model is trained by the adaptation for each epoch. Our model enables them by formulating adaptation as a generalized eigenvalue problem with one-class classification; its global optimum solution is obtained, and the solver is differentiable. We experimentally demonstrate that the proposed method achieves better performance than existing anomaly detection and few-shot learning methods on various datasets.

References in corpus (16)

• Scikit-learn: Machine Learning in Python
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Deep Learning for Anomaly Detection: A Survey
• Matching Networks for One Shot Learning
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• Efficient GAN-Based Anomaly Detection
• Linear Dimensionality Reduction: Survey, Insights, and Generalizations
• Towards a Neural Statistician
• Hierarchically Structured Meta-learning
• Generalized Inner Loop Meta-Learning
• One-Shot Generalization in Deep Generative Models
• Few-shot Autoregressive Density Estimation: Towards Learning to Learn Distributions
• The Variational Homoencoder: Learning to learn high capacity generative models from few examples
• Fisher and Kernel Fisher Discriminant Analysis: Tutorial
• Meta Learning with Relational Information for Short Sequences