# ESAD: End-to-end Deep Semi-supervised Anomaly Detection

arXiv: https://arxiv.org/abs/2012.04905 · HTML: https://mlsanity.com/paper/2012.04905/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Dec 2020.

## Built on by (most cited first, 8 of 8)

- [Registration based Few-Shot Anomaly Detection](https://mlsanity.com/paper/2207.07361/) (2022)
- [Few-Shot Anomaly Detection via Category-Agnostic Registration Learning](https://mlsanity.com/paper/2406.08810/) (2024)
- [DOC3-Deep One Class Classification using Contradictions](https://mlsanity.com/paper/2105.07636/) (2021)
- [R2-AD2: Detecting Anomalies by Analysing the Raw Gradient](https://mlsanity.com/paper/2206.10259/) (2022)
- [MELODY: Robust Semi-Supervised Hybrid Model for Entity-Level Online Anomaly Detection with Multivariate Time Series](https://mlsanity.com/paper/2401.10338/) (2024)
- [Label-Informed Outlier Detection Based on Granule Density](https://mlsanity.com/paper/2512.18774/) (2025)
- [Consistency-guided semi-supervised outlier detection in heterogeneous data using fuzzy rough sets](https://mlsanity.com/paper/2512.18977/) (2025)
- [Outlier detection in mixed-attribute data: a semi-supervised approach with fuzzy approximations and relative entropy](https://mlsanity.com/paper/2512.18978/) (2025)

## Abstract (arXiv)

This paper explores semi-supervised anomaly detection, a more practical setting for anomaly detection where a small additional set of labeled samples are provided. We propose a new KL-divergence based objective function for semi-supervised anomaly detection, and show that two factors: the mutual information between the data and latent representations, and the entropy of latent representations, constitute an integral objective function for anomaly detection. To resolve the contradiction in simultaneously optimizing the two factors, we propose a novel encoder-decoder-encoder structure, with the first encoder focusing on optimizing the mutual information and the second encoder focusing on optimizing the entropy. The two encoders are enforced to share similar encoding with a consistent constraint on their latent representations. Extensive experiments have revealed that the proposed method significantly outperforms several state-of-the-arts on multiple benchmark datasets, including medical diagnosis and several classic anomaly detection benchmarks.