New paper on cause-aware resilience for Split Computing.
- Post by: admin
- June 24, 2026
- No Comment
New paper on cause-aware resilience for Split Computing.
Another great piece of work from our group in collaboration with the CAD & Reliability Group @ PoliTO. The article “AI-Based Detection and Classification of Adversarial and Fault-Induced Threats in Split Computing,” published in the IEEE Transactions on Device and Materials Reliability, is now available on IEEE Xplore.
Link to the full paper: https://ieeexplore.ieee.org/document/11570223.
We propose a lightweight AI classifier that monitors intermediate feature maps and distinguishes between clean, hardware fault–induced, and adversarial attack–induced behavior at runtime in split computing settings.
Congratulations to the entire team: Giuseppe Esposito, Enrico Magliano, Tamer Ahmed Eltaras, Nicola Scarano, Juan David Guerrero Balaguera, Luca Mannella, Josie Esteban Rodriguez Condia, Annachiara Ruospo, Stefano Di Carlo, Marco Levorato, Alessandro Savino, Matteo Sonza Reorda.
#AI #NeuralNetworks #HardwareReliability #HardwareSecurity #SplitComputing #SafetyCriticalSystems
#TransactionJournal #TDMR
View original post on LinkedIn