Newsfeed November & December 2023

Published papers

• Mohammad Sadegh Aslanpour, Adel Nadjaran Toosi, Muhammad Aamir Cheema, Mohan Baruwal Chhetri, “faasHouse: Sustainable Serverless Edge Computing through Energy-aware Resource Scheduling”, in IEEE Transactions on Services Computing (IF – 8.1). This paper proposes an energy-aware scheduler for resource-constrained serverless edge systems.  [ePublish: EP2023-2714]
• Mohammad Sadegh Aslanpour, Adel Nadjaran Toosi, Muhammad Aamir Cheema, Mohan Baruwal Chhetri, Mohsen Amini Salehi, “Load Balancing for Heterogeneous Serverless Edge Computing: A Performance-Driven and Empirical Approach”, in Future Generation Computer Systems (IF – 7.5). This paper proposes a load-balancing approach for heterogeneous serverless edge systems based on weighted multi-objective optimisation.
  
• Monica Whitty, Nour Moustafa, Marthie Grobler. ‘Cyber security when working from home during COVID-19: Considering the human factors’. In Journal of Cybersecurity (IF 2.3).  This research examined the lives of Australian employees who moved to work from home during COVID-19. Taking a unique approach to cyber security, we sought to gain insights into the intermingling of individuals’ personal lives and technology to inform policies and educational programmes. These findings are valuable post-COVID-19, especially if the ‘new working normal’ provides options for employees to work from home.
• Saud Khan, Chandra Thapa, Salman Durrani, and Seyit Camtepe, “Access-based Lightweight Physical Layer Authentication for the Internet of Things Devices,” in IEEE Internet of Things Journal (Impact factor: 10.238, group target list), doi: 10.1109/JIOT.2023.3331362. At the dawn of 6G networks, we utilize the inherent properties of the IoT devices’ transmission model as its source for seed generation and device authentication. This paper is a big step to bring intelligence to the physical layer (i.e., RFAI) to ensure low-cost security through continuous authentication by checking the access time slots and spreading sequences of the IoT devices instead of repeatedly generating and verifying shared keys.

Publications accepted

• Hongsheng Hu, Shuo Wang, Jiamin Chang, Haonan Zhong, Ruoxi Sun, Shuang Hao, Haojin Zhu, and Minhui Xue. A Duty to Forget, a Right to be Assured? Exposing Vulnerabilities in Machine Unlearning Services. In Network and Distributed System Security Symposium (NDSS) 2024 (CORE A*). In this paper, we emphasize the challenge and vulnerabilities in current machine unlearning services, a concept vital to upholding data privacy and regulatory compliance like GDPR and CCPA. Our investigation uncovers the feasibility of “over-unlearning” in MLaaS platforms, a scenario where unlearning processes compromise more data than intended. This discovery has profound implications, highlighting a potential gap in data security and privacy in widely used MLaaS systems.
• Guoxin Sun, T. Alpcan, B. I. P. Rubinstein and S. Camtepe, “To Act or Not To Act: An Adversarial Game for Securing Vehicle Platoons,” in IEEE Transactions on Information Forensics and Security, doi: 10.1109/TIFS.2023.3320610. (group target list) This work focuses on end-to-end attack detection and mitigation approaches for operational ML-enabled systems by uniquely incorporating advancements in (adversarial) machine learning, control theory, and game theory. The work highlights a feasible architecture for future security solutions, as rule-based or ML-only approaches do not capture operational challenges well enough.

• Ruoxi Sun, Minhui Xue, Gareth Tyson, Tian Dong, Shaofeng Li, Shuo Wang, Haojin Zhu, Seyit Camtepe, and Surya Nepal, Mate! Are You Really Aware? An Explainability-Guided Testing Framework for Robustness of Malware Detectors, ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE), 2023. (SIGSOFT Distinguished Paper Award at ESEC/FSE 2023). Jason Xue has also received a Distinguished Reviewer Award at ESEC/FSE 2023 (CORE A*, top SE conference)

• David D. Nguyen, David Liebowitz, Salil Kanhere, Surya Nepal, “Multiple Hypothesis Dropout: Estimating the Parameters of Multi-Modal Output Distributions“, in AAAI Conference on Artificial Intelligence, 2024, (AAAI, CORE A*).  The paper introduces a novel multiple-output function and highlights its broad applicability across supervised and generative modelling problems.
• Jeff Lu, Dong Gong, Shuo Wang, Minhui Xue, Lina Yao, Kristen Moore. “Learning with Mixture of Prototypes for Out-of-Distribution Detection”, in ICLR: International Conference on Learning Representations , 2024, (ICLR, CORE A*). This paper introduces a new model for distance-based out-of-distribution detection that utilizes multiple prototypes for each class.

Projects

• The CIPR Mission collaboration with Google is focused and developing practical new methods and tools that will enable CI operators to assess software for vulnerabilities to reduce disruptions to Australia’s essential services for this growing threat. The contract has been signed (Google investment $500K – CSIRO investment – $625K) and has commenced. A media release is out jointly by CSIRO and Monash, regarding a paper published in the 43rd International Cryptology Conference (Crypto 2023), co-authored by Monash and Data61 researchers.  A new algorithm in the paper achieves more efficient quantum-safe zero-knowledge proof that find various security applications like securing large decentralized systems by Muhammed Esgin, Ron Steinfeld,  Dongxi Liu and Sushmita Ruj. https://www.monash.edu/news/articles/the-new-frontier-in-online-security-quantum-safe-cryptograp