Guy Amir

Papers

Total Citations

H-Index

About

Guy Amir is an emerging researcher specializing in the verification and safety of deep learning systems, with a particular focus on deep reinforcement learning (DRL) applied to robotics and autonomous navigation. His work addresses a critical challenge in modern AI: ensuring that learned policies embedded in reactive systems are reliable, correct, and safe — especially in high-stakes environments where hardware integrity and human safety are at risk. Amir's most notable contribution, "Verifying Learning-Based Robotic Navigation Systems," has garnered 18 citations since its 2023 publication, establishing him as a voice in the underexplored intersection of formal DNN verification and robotic systems. His research demonstrates that standard verification techniques can be meaningfully extended to real-world navigation policies — a significant step forward given how little prior work had tackled this problem. Complementing this, his research on constrained reinforcement learning through scenario-based programming proposes principled frameworks for training DRL agents under safety constraints, offering practical tools for deploying AI in safety-critical robotic applications. Taken together, Amir's contributions push the boundaries of trustworthy AI, making him a researcher to watch as the field increasingly demands provably safe autonomous systems.