|
You are welcome to submit your contributions to the workshop on Human-oriented Intelligent Defence Against Malware Threats, which will be held as a part of the 28th International Joint Conference on Artificial Intelligence (IJCAI) 2019. The conference will take place on the 10-16th of August 2019 in Macao, China.
Workshop webpage and detailed information: https://folk.ntnu.no/andriis/hidamt2019/
*** IMPORTANT DATES ***
May 13 (extended from Apr 12), 2019: Due date for full workshop paper submissions
June 3 (extended from May 10), 2019: Paper acceptance notification
June 14 (extended from Jun 3), 2019: Final camera-ready papers submission
Aug 10-12, 2019: Workshops and conference
*** INTRODUCTION ***
Recent cybersecurity incidents involving malware demonstrated how serious the consequences can be for both individual users and large organizations. McAfee report on malware threats shows that over four quarters of 2017 there were identified 690 millions of malware samples, which is an extreme number considering amount of manual work required to process even a tiny fraction of those. Many malware analysis across different security organizations spent hours trying to analyze and understand functionality of malware. At the same time overwhelming amount of malicious threats and malware forms cause considerable delays from the time malware has been discovered to the time a corresponding efficient signature was created. Moreover, the malware infection is no longer limited to personal computers, but now also hits such components as Internet of Things and Industrial Control Systems, which were previously unaffected and the cybersecirty impact was underestimated. From before Machine Learning and Computational Intelligence have demonstrated advantages of application in cybersecurity-related tasks. In particular, many researchers have been employing such techniques to mitigate obfuscation, polymorphous and encryption while building intelligent malware detection mechanisms. Intelligent malware analysis and detection is an emerging topic of cybersecurity that has to go in line with advancement of malware developers and consistent presence of zero-day attacks. Our focus is not only to build and effective Machine Learning-based malware protection, but also comprise models that are to be understood by human experts. Therefore, we believe that Machine Learning-aided human-oriented approaches will ensure timely response to malware threats. Moreover, those can serve as a stepping stone in faster and more efficient analysis of novel malware as well as similarity-based identification of adversarial attacks on Machine Learning.
*** PROPOSED TOPICS ***
Note that the topics are not limited to this proposed list.
1. Automated pre-processing phase
- Efficient features identification and construction
- Novel approaches for malware categorization
- Human-understandable characterization of malware
- Indicators of Compromise as successful identification tool
- Information Fusion and Open Threats Intelligence
2. Advanced computational methods
- Deep Learning models
- Similarity-based analysis to avoid evasion
- Big Data-oriented optimization of detection
- Hybrid Intelligence and Soft Computing
- Secure and robust models to avoid adversarial attacks
3. Combating malware in a wild
- End-point implementations
- Novel malware collection and sharing platforms
- Applications in Decision Support Systems
- Human reasoning in Machine Learning-aided malware detection
- Real-time defence and online learning
- Explainable rules derived from train Machine Learning models
*** PROGRAM CO-CHAIRS ***
Andrii Shalaginov, Norwegian University of Science and Technology
Geir Olav Dyrkolbotn, Center for Cyber and Information Security
Sergii Banin, Norwegian University of Science and Technology
Ali Dehghantanha, University of Guelph
Katrin Franke, Norwegian University of Science and Technology
*** PROGRAM COMMITTEE ***
Olaf M. Maennel (Tallinn University of Technology)
Asif Iqbal (KTH Royal Institute of Technology)
Oleksandr Semeniuta (Norwegian University of Science and Technology)
Mamoun Alazab (Charles Darwin University)
Vasileios Mavroeidis (University of Oslo)
Sreyasee Das Bhattacharjee (University of North Carolina at Charlotte)
Igor Kotsiuba (Pukhov Institute for modeling in Energy Engineering)
Mark Scanlon (University College Dublin)
Piotr Andrzej Kowalski (AGH University of Science and Technology)
Reza Parizi (Kennesaw State University)
Mohammad Hamoudeh (Manchester Metropolitan University)
Gregory Epiphaniou (University of Wolverhampton)
Bojan Kolosnjaji (Technical University of Munich)
Shih-Chieh Su (Microsoft)
|
