Vishnu Lokhande

Vishnu Lokhande is a tenure-track Assistant Professor in the Department of Computer Science and Engineering at the University at Buffalo, SUNY, with a secondary affiliation at the Institute of Artificial Intelligence and Data Science (IAD). His research focuses on enhancing the reliability and robustness of deep learning models through innovative applications of optimization theory.

• PhD in Computer Science from University of Wisconsin-Madison (2023)
• BTech in Electrical Engineering from IIT Kanpur (2016)

Dr. Lokhande's research spans Computer Vision, Machine Learning, Deep Learning, and Mathematical Optimization, with particular emphasis on constrained optimization methods for training deep neural networks. His work bridges theoretical foundations from statistics and mathematics with practical applications in vision and learning systems.

His publication portfolio demonstrates significant contributions to constrained deep learning, particularly through his work on Conditional Gradient methods, which offer theoretical guarantees for model training while maintaining practical efficiency. Additional research areas include EEG signal processing for brain-computer interfaces, theoretical analysis of neural network convergence, and fairness-aware optimization techniques.

During his doctoral studies at Wisconsin, Dr. Lokhande completed internships at both Google Research (twice) and Microsoft Research, gaining valuable industry experience that informs his applied research approach. He is actively recruiting PhD and MS students at UB and remains open to industrial collaborations as a consultant or part-time researcher.