Mahyar Fazlyab

Mahyar Fazlyab is an Assistant Professor in the Department of Electrical and Computer Engineering at Johns Hopkins University (JHU), affiliated with the Data Science and AI Institute and the Mathematical Institute for Data Science (MINDS). He holds a Ph.D. in Electrical and Systems Engineering from the University of Pennsylvania (2018), with a dual M.A. in Statistics from the Wharton School. His research focuses on the intersection of optimization, control, and machine learning, particularly the safety and stability of learning-enabled autonomous systems. He previously served as an Assistant Research Professor at MINDS and a postdoctoral fellow at UPenn's ESE Department.

Education:

• Ph.D. in Electrical and Systems Engineering, University of Pennsylvania (2013–2018)
• M.A. in Statistics, Wharton School, University of Pennsylvania (2016–2018)
• M.Sc. & B.Sc. in Mechanical Engineering, Sharif University of Technology (2005–2013)

His research interests include machine learning for optimization, control theory, and designing algorithms for AI applications. Key areas are safety verification of neural network systems, robust optimization, and stability analysis. He has received the Joseph and Rosaline Wolf Best Doctoral Dissertation Award (2019).

Research Trends: His recent work emphasizes provable guarantees for neural networks, safe autonomous systems, and bilevel optimization. Publications span topics like certified robustness, physics-informed learning, and reachability analysis of neural-controlled systems.

Awards:

• Joseph and Rosaline Wolf Best Doctoral Dissertation Award (2019)

Advising & Grants: He mentors Ph.D. students with strong mathematical and programming backgrounds. Research is supported by grants focusing on AI safety and control systems. His lab actively explores interdisciplinary applications of optimization and machine learning.

Labs/Teams: Active member of JHU's Data Science and AI Institute and MINDS, collaborating on foundational AI research and safety-critical systems development.