Afsaneh Koohestani

Papers

Total Citations

H-Index

About

Afsaneh Koohestani is a researcher at the forefront of autonomous systems and deep imitation learning, with a focus on bridging the gap between theoretical advances and real-world deployment. Her work centers on evaluating how neural network architectures impact the performance of imitation learning in autonomous driving, a critical area where safety and reliability are paramount. In her most-cited study (2019, 19 citations), she systematically analyzed deep convolutional neural network designs for policy learning, providing key insights into how architectural choices—such as network depth, filter sizes, and skip connections—affect driving behavior and task success. This research has informed the development of more robust and efficient autonomous driving systems, helping to move imitation learning from controlled simulations toward practical applications. Koohestani’s contributions are particularly valuable for students and engineers working on end-to-end driving models, as she offers empirical guidance on model selection and training. Her work continues to shape the dialogue on how to build trustworthy AI for safety-critical environments, making her a notable voice in the autonomous driving research community.