Neural Transformation Learning for Deep Anomaly Detection Beyond Images
arXiv:2103.16440
Abstract
Data transformations (e.g. rotations, reflections, and cropping) play an important role in self-supervised learning. Typically, images are transformed into different views, and neural networks trained on tasks involving these views produce useful feature representations for downstream tasks, including anomaly detection. However, for anomaly detection beyond image data, it is often unclear which transformations to use. Here we present a simple end-to-end procedure for anomaly detection with learnable transformations. The key idea is to embed the transformed data into a semantic space such that the transformed data still resemble their untransformed form, while different transformations are easily distinguishable. Extensive experiments on time series demonstrate that our proposed method outperforms existing approaches in the one-vs.-rest setting and is competitive in the more challenging n-vs.-rest anomaly detection task. On tabular datasets from the medical and cyber-security domains, our method learns domain-specific transformations and detects anomalies more accurately than previous work.
References in corpus (20)
- Learning deep representations by mutual information estimation and maximization
- Unsupervised Representation Learning by Predicting Image Rotations
- AutoAugment: Learning Augmentation Policies from Data
- Learning Representations by Maximizing Mutual Information Across Views
- Benchmarking Multivariate Time Series Classification Algorithms
- Deep Anomaly Detection with Outlier Exposure
- Deep Anomaly Detection Using Geometric Transformations
- LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection
- On Mutual Information Maximization for Representation Learning
- CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances
- Learning to Compose Domain-Specific Transformations for Data Augmentation
- Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
- Unsupervised Detection of Lesions in Brain MRI using constrained adversarial auto-encoders
- A Bayesian Data Augmentation Approach for Learning Deep Models
- Classification-Based Anomaly Detection for General Data
- The UEA multivariate time series classification archive, 2018
- Adversarial AutoAugment
- Learning perturbation sets for robust machine learning
- Deep Semi-Supervised Anomaly Detection
- Extreme Classification via Adversarial Softmax Approximation