Research
Machine learning for Computer-Aided Design
We target to develop a machine learning framework for fault recovery in parallel and distributed systems. A faulty computing element's task can be migrated to other nodes to allow fault-tolerance computing.
We have developed both Max-Flow Min-cut adaptation and Genetic Algorithm for this flow.
Selected publications:
Khanh N. Dang, Akram Ben Ahmed, Fakhrul Zaman Rokhani, Abderazek Ben Abdallah, and Xuan-Tu Tran, ‘‘A thermal distribution, lifetime reliability prediction and spare TSV insertion platform for stacking 3D NoCs’’, 2020 International Conference On Advanced Technologies For Communications (ATC) (accepted)
Patents:
A. Ben Abdallah, Huakun Huang, Khanh N. Dang, Jiangning Song, ‘‘ＡＩプロセッサ (AI Processor)’’, 特願2020-194733, Japan patent, (under review)