Summary of Pefad: a Parameter-efficient Federated Framework For Time Series Anomaly Detection, by Ronghui Xu et al.
PeFAD: A Parameter-Efficient Federated Framework for Time Series Anomaly Detection
by Ronghui Xu, Hao Miao, Senzhang Wang, Philip S. Yu, Jianxin Wang
First submitted to arxiv on: 4 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Databases (cs.DB); Distributed, Parallel, and Cluster Computing (cs.DC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Parameter-efficient Federated Anomaly Detection framework, named PeFAD, addresses the challenge of decentralized time series anomaly detection in a privacy-preserving manner. By employing pre-trained language models as local client models and fine-tuning small-scale parameters, PeFAD reduces communication overhead and local model adaptation costs. The framework also employs an anomaly-driven mask selection strategy to mitigate the impact of neglected anomalies during training. Additionally, knowledge distillation is used on synthetic privacy-preserving datasets shared by all clients to address data heterogeneity issues. Experimental results on four real-world datasets demonstrate that PeFAD outperforms existing state-of-the-art baselines by up to 28.74%. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary PeFAD is a new way to find unusual patterns in time series data collected from many devices. This is important because this type of data can help us understand and predict many real-world phenomena. The problem is that collecting all the data in one place is not practical, so we need a solution that works with decentralized data. PeFAD uses pre-trained language models to find anomalies in the data and only requires small updates from each device. This makes it efficient and private. |
Keywords
» Artificial intelligence » Anomaly detection » Fine tuning » Knowledge distillation » Mask » Parameter efficient » Time series
