상세정보
학술지
Stabilizing Adversarially Learned One-Class Novelty Detection Using Pseudo Anomalies
- 발행일
- 202209
- 출처
- IEEE Transactions on Image Processing, v.31, pp.5963-5975
- ISSN
- 1057-7149
- 출판사
- IEEE
- 협약과제
- 22HS1300, 불확실한 지도 기반 실내ㆍ외 환경에서 최종 목적지까지 이동로봇을 가이드할 수 있는 AI 기술 개발, 이재영
- 초록
- Recently, anomaly scores have been formulated using reconstruction loss of the adversarially learned generators and/or classification loss of discriminators. Unavailability of anomaly examples in the training data makes optimization of such networks challenging. Attributed to the adversarial training, performance of such models fluctuates drastically with each training step, making it difficult to halt the training at an optimal point. In the current study, we propose a robust anomaly detection framework that overcomes such instability by transforming the fundamental role of the discriminator from identifying real vs. fake data to distinguishing good vs. bad quality reconstructions. For this purpose, we propose a method that utilizes the current state as well as an old state of the same generator to create good and bad quality reconstruction examples. The discriminator is trained on these examples to detect the subtle distortions that are often present in the reconstructions of anomalous data. In addition, we propose an efficient generic criterion to stop the training of our model, ensuring elevated performance. Extensive experiments performed on six datasets across multiple domains including image and video based anomaly detection, medical diagnosis, and network security, have demonstrated excellent performance of our approach.
- KSP 제안 키워드
- Adversarial Training, Current state, Medical diagnosis, Multiple domains, anomaly detection framework, excellent performance, network security, novelty detection, training data, video based