Gavin Brown
Papers
1
Total Citations
77
H-Index
1
About
Gavin Brown is a leading researcher at the intersection of trustworthy machine learning and robotics, with a focus on the safety and reliability of deep neural networks in physical systems. His most-cited work, "Is Deep Learning Safe for Robot Vision? Adversarial Examples against the iCub Humanoid" (2018, 77 citations), is a landmark study that exposed critical vulnerabilities in deep learning when deployed on robotic platforms. By demonstrating that adversarial examples—imperceptible perturbations to images—can successfully fool the iCub humanoid robot’s vision system, Brown’s research bridged the gap between theoretical adversarial machine learning and real-world safety concerns. This work has been instrumental in raising awareness about the risks of deploying deep learning in safety-critical applications, such as autonomous navigation and human-robot interaction. Brown’s contributions have helped shape the emerging field of adversarial robustness in robotics, influencing subsequent research on defensive mechanisms and certification methods. His work is essential reading for students and researchers interested in building AI systems that are not only powerful but also trustworthy and resilient in the face of malicious inputs.
Research Focus
Key Achievements
Top Papers
- 1