Spot The Odd One Out: Regularized Complete Cycle Consistent Anomaly Detector GAN

16 Apr 2023  ยท  Zahra Dehghanian, Saeed Saravani, Maryam Amirmazlaghani, Mohammad Rahmati ยท

This study presents an adversarial method for anomaly detection in real-world applications, leveraging the power of generative adversarial neural networks (GANs) through cycle consistency in reconstruction error. Previous methods suffer from the high variance between class-wise accuracy which leads to not being applicable for all types of anomalies. The proposed method named RCALAD tries to solve this problem by introducing a novel discriminator to the structure, which results in a more efficient training process. Additionally, RCALAD employs a supplementary distribution in the input space to steer reconstructions toward the normal data distribution, effectively separating anomalous samples from their reconstructions and facilitating more accurate anomaly detection. To further enhance the performance of the model, two novel anomaly scores are introduced. The proposed model has been thoroughly evaluated through extensive experiments on six various datasets, yielding results that demonstrate its superiority over existing state-of-the-art models. The code is readily available to the research community at https://github.com/zahraDehghanian97/RCALAD.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Anomaly Detection CIFAR-10 RCALAD Mean AUC 65.7 # 1
Anomaly Detection KDD Cup 1999 RCALAD F1-Score 95.4 # 1
Anomaly Detection MIT-BIH Arrhythmia Database RCALAD F1 score 60.6 # 1
Anomaly Detection Musk v1 RCALAD F1-Score 63.1 # 1
Anomaly Detection SVHN RCALAD Mean AUC 57.7 # 1
Anomaly Detection Thyroid RCALAD F1-Score 52.6 # 1

Methods


No methods listed for this paper. Add relevant methods here