Darren Cofer
Papers
1
Total Citations
6
H-Index
1
About
Darren Cofer is a leading figure in the formal verification and safety assurance of cyber-physical systems, with a particular focus on the rigorous testing and certification of neural networks used in safety-critical applications. His work addresses the pressing challenge of ensuring that deep neural networks (DNNs)—now integral to autonomous vehicles, medical diagnostics, and industrial robotics—can be trusted to operate without catastrophic failure. Cofer’s most-cited paper, “Input Prioritization for Testing Neural Networks” (2019), introduces a method for efficiently identifying high-risk inputs that are most likely to expose faults, thereby making the testing process both more effective and scalable. This contribution is vital for certifying systems where failures could lead to loss of life or property. With over 6 citations on this work alone, Cofer’s research has helped bridge the gap between traditional formal methods and modern machine learning, enabling engineers to systematically validate neural network behavior. His broader portfolio includes pioneering work on model-based development and architectural safety patterns, making him a key voice in the movement toward verifiable autonomy.
Research Focus
Key Achievements
Top Papers
- 1Input Prioritization for Testing Neural Networks6 citations · 2019