Fin-Fed-OD: Federated Outlier Detection on Financial Tabular Data

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Herurkar, Dayananda, Palacio, Sebastian, Anwar, Ahmed, Hees, Joern, Dengel, Andreas
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917647961030656
author Herurkar, Dayananda
Palacio, Sebastian
Anwar, Ahmed
Hees, Joern
Dengel, Andreas
author_facet Herurkar, Dayananda
Palacio, Sebastian
Anwar, Ahmed
Hees, Joern
Dengel, Andreas
contents Anomaly detection in real-world scenarios poses challenges due to dynamic and often unknown anomaly distributions, requiring robust methods that operate under an open-world assumption. This challenge is exacerbated in practical settings, where models are employed by private organizations, precluding data sharing due to privacy and competitive concerns. Despite potential benefits, the sharing of anomaly information across organizations is restricted. This paper addresses the question of enhancing outlier detection within individual organizations without compromising data confidentiality. We propose a novel method leveraging representation learning and federated learning techniques to improve the detection of unknown anomalies. Specifically, our approach utilizes latent representations obtained from client-owned autoencoders to refine the decision boundary of inliers. Notably, only model parameters are shared between organizations, preserving data privacy. The efficacy of our proposed method is evaluated on two standard financial tabular datasets and an image dataset for anomaly detection in a distributed setting. The results demonstrate a strong improvement in the classification of unknown outliers during the inference phase for each organization's model.
format Preprint
id arxiv_https___arxiv_org_abs_2404_14933
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fin-Fed-OD: Federated Outlier Detection on Financial Tabular Data
Herurkar, Dayananda
Palacio, Sebastian
Anwar, Ahmed
Hees, Joern
Dengel, Andreas
Machine Learning
Artificial Intelligence
Anomaly detection in real-world scenarios poses challenges due to dynamic and often unknown anomaly distributions, requiring robust methods that operate under an open-world assumption. This challenge is exacerbated in practical settings, where models are employed by private organizations, precluding data sharing due to privacy and competitive concerns. Despite potential benefits, the sharing of anomaly information across organizations is restricted. This paper addresses the question of enhancing outlier detection within individual organizations without compromising data confidentiality. We propose a novel method leveraging representation learning and federated learning techniques to improve the detection of unknown anomalies. Specifically, our approach utilizes latent representations obtained from client-owned autoencoders to refine the decision boundary of inliers. Notably, only model parameters are shared between organizations, preserving data privacy. The efficacy of our proposed method is evaluated on two standard financial tabular datasets and an image dataset for anomaly detection in a distributed setting. The results demonstrate a strong improvement in the classification of unknown outliers during the inference phase for each organization's model.
title Fin-Fed-OD: Federated Outlier Detection on Financial Tabular Data
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2404.14933