Olivier Sentieys
Papers
1
Total Citations
12
H-Index
1
About
Olivier Sentieys is a leading researcher in energy-efficient computing, approximate computing, and fault-tolerant systems for deep neural networks (DNNs). His work bridges the gap between hardware reliability and machine learning, addressing critical challenges in deploying DNNs in safety-critical domains like robotics, aerospace, and autonomous driving. A standout contribution is the development of **harDNNing**, a machine-learning-based framework for assessing and protecting DNNs against hardware faults. This work, published in 2023 and already garnering 12 citations, exemplifies his approach of combining algorithmic resilience with hardware-aware design. Sentieys has also made foundational contributions to approximate computing, where he pioneered techniques to trade off computational accuracy for significant energy savings, and to low-power embedded systems design. His research has been widely recognized, with his most-cited papers accumulating hundreds of citations, reflecting their impact on both academia and industry. Through his leadership at INRIA and the University of Rennes, Sentieys continues to shape the future of reliable, energy-efficient AI systems, making his work essential reading for students and researchers interested in the intersection of hardware, software, and machine learning.
Research Focus
Key Achievements
Top Papers
- 1