A Reliable Framework for Human-in-the-Loop Anomaly Detection in Time Series

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Deng, Ziquan, Xuan, Xiwei, Ma, Kwan-Liu, Kong, Zhaodan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911019137236992
author Deng, Ziquan
Xuan, Xiwei
Ma, Kwan-Liu
Kong, Zhaodan
author_facet Deng, Ziquan
Xuan, Xiwei
Ma, Kwan-Liu
Kong, Zhaodan
contents Time series anomaly detection is a critical machine learning task for numerous applications, such as finance, healthcare, and industrial systems. However, even high-performing models may exhibit potential issues such as biases, leading to unreliable outcomes and misplaced confidence. While model explanation techniques, particularly visual explanations, offer valuable insights by elucidating model attributions of their decision, many limitations still exist -- They are primarily instance-based and not scalable across the dataset, and they provide one-directional information from the model to the human side, lacking a mechanism for users to address detected issues. To fulfill these gaps, we introduce HILAD, a novel framework designed to foster a dynamic and bidirectional collaboration between humans and AI for enhancing anomaly detection models in time series. Through our visual interface, HILAD empowers domain experts to detect, interpret, and correct unexpected model behaviors at scale. Our evaluation through user studies with two models and three time series datasets demonstrates the effectiveness of HILAD, which fosters a deeper model understanding, immediate corrective actions, and model reliability enhancement.
format Preprint
id arxiv_https___arxiv_org_abs_2405_03234
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Reliable Framework for Human-in-the-Loop Anomaly Detection in Time Series
Deng, Ziquan
Xuan, Xiwei
Ma, Kwan-Liu
Kong, Zhaodan
Human-Computer Interaction
Machine Learning
Time series anomaly detection is a critical machine learning task for numerous applications, such as finance, healthcare, and industrial systems. However, even high-performing models may exhibit potential issues such as biases, leading to unreliable outcomes and misplaced confidence. While model explanation techniques, particularly visual explanations, offer valuable insights by elucidating model attributions of their decision, many limitations still exist -- They are primarily instance-based and not scalable across the dataset, and they provide one-directional information from the model to the human side, lacking a mechanism for users to address detected issues. To fulfill these gaps, we introduce HILAD, a novel framework designed to foster a dynamic and bidirectional collaboration between humans and AI for enhancing anomaly detection models in time series. Through our visual interface, HILAD empowers domain experts to detect, interpret, and correct unexpected model behaviors at scale. Our evaluation through user studies with two models and three time series datasets demonstrates the effectiveness of HILAD, which fosters a deeper model understanding, immediate corrective actions, and model reliability enhancement.
title A Reliable Framework for Human-in-the-Loop Anomaly Detection in Time Series
topic Human-Computer Interaction
Machine Learning
url https://arxiv.org/abs/2405.03234