---
title: Explainable Autoencoder Anomaly Detection
url: https://www.ml-quant.com/papers/ssrn/4819146/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: SSRN 4819146
source_url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4819146
featured: 2024-05-08
citations: unknown
topic: ML & AI Methods
---


# Explainable Autoencoder Anomaly Detection

The suggested model uses an explainable variational autoencoder to detect anomalies in multivariate time series data, overcoming issues of large data size, unknown anomalies, and unclear deep learning detection methods.

- Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4819146
- Identifier: SSRN 4819146
- Released: 2024-05-07
- First featured: Quant Letter No. 48 (2024-05-08): https://www.ml-quant.com/issues/2024-05-08/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: ML & AI Methods

## Related

- [Identifying regions of importance in wall-bounded turbulence through explainable deep learning](https://www.ml-quant.com/papers/arxiv/2302.01250/): A deep-learning method has been used to analyze energy structures in turbulence, finding that the most significant structures aren't always the ones contributing most to Reynolds shear stress.
- [Year-over-Year Developments in Financial Fraud Detection via Deep Learning: A Systematic Literature Review](https://www.ml-quant.com/papers/arxiv/2502.00201/): The study examines progress in deep learning methods for detecting financial fraud, reviewing 57 studies from 2019 to 2024, and discusses challenges and opportunities such as data privacy, feature engineering, and model interpretability.
- [Generative AI enables medical image segmentation in ultra low-data regimes](https://www.ml-quant.com/papers/arxiv/2408.17421/): A new deep learning framework generates high-quality medical images and segmentation masks, improving model performance in data-limited situations and reducing the need for extensive training data.
- [A Copula-Guided In-Model Interpretable Neural Network for Change Detection in Heterogeneous Remote Sensing Images](https://www.ml-quant.com/papers/arxiv/2303.17448/): NN-Copula-CD, a new method for detecting changes in heterogeneous remote sensing images, combines copula theory and deep neural networks, improving disaster monitoring and land-use management.
- [Historical calibration of SVJD models with deep learning](https://www.ml-quant.com/papers/ssrn/4650097/): The paper suggests using deep neural networks to calibrate parameters of Stochastic Volatility Jump Diffusion models, proving to be more accurate, robust, and faster than other methods.
- [Explainable representation learning of small quantum states](https://www.ml-quant.com/papers/arxiv/2306.05694/): The research examines the learned representation of a generative model trained on two-qubit density matrices, showing a direct correlation with the entanglement measure concurrence, providing insights into machine learning of quantum states.