Rudy Bunel
Papers
1
Total Citations
7
H-Index
1
About
Rudy Bunel is a leading researcher in the formal verification and safety certification of neural networks, with a particular focus on bridging the gap between machine learning and control theory. His work addresses the critical challenge of ensuring that neural-network-controlled systems—such as autonomous vehicles—operate safely, even in complex, real-world environments. Bunel is best known for pioneering scalable verification techniques, including branch-and-bound methods for neural network robustness and the DRIP (Domain Refinement Iteration with Polytopes) framework for backward reachability analysis. DRIP, published in 2023 and already garnering 7 citations, provides a novel approach to certifying collision avoidance guarantees by computing exact backward reachable sets for neural feedback loops, overcoming the non-invertibility of neural networks. His contributions have been instrumental in advancing the theoretical foundations of neural network verification, enabling more reliable deployment of AI in safety-critical applications. Bunel’s work continues to influence both academic research and practical engineering, making him a key figure in the quest for trustworthy autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1