GANFL: A log anomaly detection method based on collaborative optimization of federated learning and generative adversarial networks

Longxin Yao, Xuanran Li, Mingzhe Li, Bo Zhang
Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, PMLR 278:760-767, 2025.

Abstract

With the rapid development of information technology, the amount of data is growing explosively. Enterprises and society have an increasing demand for data storage, processing and analysis. Data centers have emerged as the times require. They can centrally manage massive amounts of data, provide efficient computing and storage capabilities, meet the high requirements of different industries for data processing, and ensure data security and reliability. In data centers, numerous devices, systems and applications continuously generate a large number of logs during operation. These logs record the activities and status information at all levels of the data center, including the operating status of the server, the traffic of network devices, and the operation records of applications. Log anomalies refer to the presence of records that do not conform to normal patterns or expected content in the log files that record the operating system’s own operating events. Log analysis can help developers quickly locate the source of the fault. By analyzing the log data, they can determine the device where the fault occurred and the cause of the fault. At the same time, they can also conduct advance analysis based on the existing log data to discover potential problems.In this paper, the method of co-optimization of GAN and federated learning is adopted, which not only solves the problem of data silos, but also solves the problem of insufficient data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v278-yao25a, title = {GANFL: A log anomaly detection method based on collaborative optimization of federated learning and generative adversarial networks}, author = {Yao, Longxin and Li, Xuanran and Li, Mingzhe and Zhang, Bo}, booktitle = {Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing}, pages = {760--767}, year = {2025}, editor = {Zeng, Nianyin and Pachori, Ram Bilas and Wang, Dongshu}, volume = {278}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v278/main/assets/yao25a/yao25a.pdf}, url = {https://proceedings.mlr.press/v278/yao25a.html}, abstract = { With the rapid development of information technology, the amount of data is growing explosively. Enterprises and society have an increasing demand for data storage, processing and analysis. Data centers have emerged as the times require. They can centrally manage massive amounts of data, provide efficient computing and storage capabilities, meet the high requirements of different industries for data processing, and ensure data security and reliability. In data centers, numerous devices, systems and applications continuously generate a large number of logs during operation. These logs record the activities and status information at all levels of the data center, including the operating status of the server, the traffic of network devices, and the operation records of applications. Log anomalies refer to the presence of records that do not conform to normal patterns or expected content in the log files that record the operating system’s own operating events. Log analysis can help developers quickly locate the source of the fault. By analyzing the log data, they can determine the device where the fault occurred and the cause of the fault. At the same time, they can also conduct advance analysis based on the existing log data to discover potential problems.In this paper, the method of co-optimization of GAN and federated learning is adopted, which not only solves the problem of data silos, but also solves the problem of insufficient data.} }
Endnote
%0 Conference Paper %T GANFL: A log anomaly detection method based on collaborative optimization of federated learning and generative adversarial networks %A Longxin Yao %A Xuanran Li %A Mingzhe Li %A Bo Zhang %B Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing %C Proceedings of Machine Learning Research %D 2025 %E Nianyin Zeng %E Ram Bilas Pachori %E Dongshu Wang %F pmlr-v278-yao25a %I PMLR %P 760--767 %U https://proceedings.mlr.press/v278/yao25a.html %V 278 %X With the rapid development of information technology, the amount of data is growing explosively. Enterprises and society have an increasing demand for data storage, processing and analysis. Data centers have emerged as the times require. They can centrally manage massive amounts of data, provide efficient computing and storage capabilities, meet the high requirements of different industries for data processing, and ensure data security and reliability. In data centers, numerous devices, systems and applications continuously generate a large number of logs during operation. These logs record the activities and status information at all levels of the data center, including the operating status of the server, the traffic of network devices, and the operation records of applications. Log anomalies refer to the presence of records that do not conform to normal patterns or expected content in the log files that record the operating system’s own operating events. Log analysis can help developers quickly locate the source of the fault. By analyzing the log data, they can determine the device where the fault occurred and the cause of the fault. At the same time, they can also conduct advance analysis based on the existing log data to discover potential problems.In this paper, the method of co-optimization of GAN and federated learning is adopted, which not only solves the problem of data silos, but also solves the problem of insufficient data.
APA
Yao, L., Li, X., Li, M. & Zhang, B.. (2025). GANFL: A log anomaly detection method based on collaborative optimization of federated learning and generative adversarial networks. Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, in Proceedings of Machine Learning Research 278:760-767 Available from https://proceedings.mlr.press/v278/yao25a.html.

Related Material