Anomaly Detection via Reverse Distillation from One-Class Embedding

🤖 Plain-English Summary

Knowledge distillation (KD) achieves promising results on the challenging problem of unsupervised anomaly detection (AD). The obtained compact embedding effectively preserves essential information on normal patterns, but aban-dons anomaly perturbations.

🔑 Key Findings

• The representation discrepancy of anomalies in the teacher-student (T-S) model provides essential evidence for AD.
• However, using similar or identical architectures to build the teacher and student models in previous studies hinders the diversity of anomalous representations.
• To tackle this problem, we propose a novel T-S model consisting of a teacher encoder and a student decoder and introduce a simple yet effective “reverse distillation” paradigm accordingly.

🔗 Related Resources