New Research by TSYS School Computer Scientist Riduan Abid Examines Use of Split Neural Networks in Edge Computing

Cloud computing is a critical component in the success of 5G and 6G networks, particularly given the computation-intensive nature of emerging applications. Despite all it advantages, cloud computing faces limitations in meeting the strict latency and bandwidth requirements of applications such as eHealth and automotive systems. To overcome these limitations, edge computing has emerged as a novel paradigm that bring computation closer to the user. Moreover, intelligent tasks related to deep learning demand more memory and processing power than edge devices can handle. To address these challenges, methods like quantization, pruning, and distributed inference have been proposed. A new study by TSYS School computer scientist Riduan Abid, Salmane Douch, Khalid Zine-Dine and Driss Bouzidi of Mohammed V University in Rabat, and Driss Benhaddou of Alfaisal University examines a promising approach for running deep learning models at the edge that employs split neural networks. Split neural networks feature a neural network architecture with multiple early exit points, allowing the model to make confident decisions at earlier layers without processing the entire network. This not only reduces memory and computational demands but it also makes split neural networks well-suited for edge computing applications. As the use of split neural networks expands, ensuring their safety—particularly their robustness to perturbations—becomes crucial for deployment in safety-critical scenarios. Abid et al.'s new study, set to appear in a future issue of IEEE Access, presents the first in-depth study on the robustness of split Edge Cloud neural networks. In doing so it reviews state-of-the-art robustness certification techniques and evaluate split neural networks robustness using the auto_LiRPA and Auto Attack libraries, comparing them to standard neural networks. The results demonstrate that split neural networks reduce average inference time by 75‘% and certify four to 10 times more images as robust, while improving overall robustness accuracy by 1% to 10%.

Comments

Four Turner College Faculty Recognized for Outstanding Teaching

CSU Head Women's Soccer Coach Jay Entlich recently released a list of CSU faculty who have been chosen by a player as a member of the CSU faculty who has impacted the player in a positive way along their journey at CSU. Four Turner College faculty were included on the list, along with the player who nominated each. Management professor Phil Bryant was named by Sophia Leal , a freshman midfielder from Oxford, Georgia. Sophia attended Eastside High School and was a two-time all-region selection during her high school career. Through the first 10 games of 2024, she has scored one goal and recorded three assists. Next, management professor John Finley was named by Lizz Forshaw , a graduate student forward from Stockton, England. Lizz, who attended IMG Academy in south Florida, has scored four goals and recorded four assists this season. During her senior year in 2023, she scored three goals and recorded two assists. As a junior in 2022, Lizz scored three goals and recorded fo

Turner College Grad Publishes Academic Study on Corporate Culture and Strategy

Former Turner College student Tamara Todorova , now an associate professor of economics at American University in Bulgaria (AUB), recently published a study on corporate culture and strategy. Todorova earned an MBA from the Turner College in 1996 and then went on to earn a doctorate in international economics from the University of Economics - Varna in 2001. She has been on the faculty at AUB since August of 2000. Todorova's study, which appears in the current issue of the International Journal of Business Performance Management , investigates how corporate culture helps to economize on the transaction costs of internal organization. As she explains, the dimensions of corporate culture that assist in this task include increasing trust and reducing intrafirm opportunism. Todorova's study demonstrates that setting common goals and a common direction reduces the sizeable costs of internal organization. Tamara's prior research appears in Economics of Transition , International

Hammer joins Turner College Finance Faculty

The Turner College's Department of Accounting and Finance welcomes Melissa Ingle Hammer , who joins the group as an adjunct faculty in finance. Hammer graduated from the Turner College in 2005, after earning a BBA in finance. She later went on to earn an MBA from Auburn University, a Specialist Certificate in Conduct Risk from the University of Manchester, a Certificate in Team Culture from the University of Pennsylvania, and a Certificate in Executive Presence from Cornell University. Hammer is the SVP of Fair Lending Product Management with RiskExec at Asurity. She previously worked as Compliance Manager of Fair Lending at Synovus, a senior regulatory consultant with Wolters Kluwer, and as VP of Compliance at TD. This semester Hammer will be teaching FINC 3135, Financial Institutions and Technologies. Turner Business welcomes Melissa and wishes her a great first semester in the Turner College.

Turner Business

Search This Blog