@inproceedings{oai:kansai-u.repo.nii.ac.jp:00013560,
 author = {Shimizu, Koki and Kumai, Yuya and Motonaka, Kimiko and Kimura, Tomotaka and Hirata, Kouji},
 month = {2019-11, 2019-10-01},
 note = {Recently, machine learning technologies have dramatically evolved. Accordingly, the concept of self-evolving botnets has been introduced, which discover vulnerabilities of hosts by distributed machine learning using the computational resources of infected hosts, and infect other hosts by attacks using the discovered vulnerabilities. The infectability of the self-evolving botnets is too strong compared with conventional botnets, so that such new botnets will become the serious threat to future network society including 5G and IoT environments. In this paper, we consider a volunteer model that discovers unknown vulnerabilities earlier than self-evolving botnets by distributed computing using volunteer hosts’ resources and repairs the vulnerabilities. We propose deterministic modeling for the volunteer model. Through numerical calculations, we evaluate the performance of the volunteer model against self-evolving botnets., This is a product of research which was financially supported by the Kansai University Fund for Supporting Young Scholars, 2018, "Design of anti-malware systems against future malware evolution". This research was partially supported by The Telecommunications Advancement Foundation, Japan., Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2019), November 18-21, 2019, Lanzhou, China},
 title = {Evaluation of countermeasure against future malware evolution with deterministic modeling},
 year = {}
}