Paolo Rech
Papers
3
Total Citations
40
H-Index
2
About
Paolo Rech is a leading researcher at the intersection of high-performance computing and system reliability, with a primary focus on the dependability of Graphics Processing Units (GPUs) and Edge AI accelerators in safety-critical environments. His work is pivotal in understanding how hardware faults—particularly those induced by neutron radiation—affect the execution of modern Artificial Intelligence workloads. Rech’s major contributions include pioneering a multi-level methodology to evaluate the impact of permanent GPU faults on Convolutional Neural Networks (CNNs), a study that has garnered 32 citations and is essential for long-lifetime applications like automotive and robotics. He has also advanced the field by measuring the neutron sensitivity of Deep Reinforcement Learning policies on Google’s Coral Edge TPU, directly informing the design of reliable autonomous systems that interact with humans. Most recently, his research has extended to space robotics, evaluating the reliability of Vision Transformers for planet exploration. With over 40 citations across his most influential works, Rech’s systematic approach to fault injection and resilience analysis is shaping the next generation of robust, AI-powered autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3