Gaurang Sriramanan

I’m a third-year Computer Science PhD student at The University of Maryland (UMD), College Park. I’m very fortunate to be advised by Prof. Soheil Feizi, at the Center for Machine Learning. My primary interest involves exploring and characterizing various vulnerabilities of Deep Neural Networks and developing training methodologies that enhance their robustness and reliability. Despite the remarkable performance of current Machine Learning models on standard data, their adaptability under distribution shifts and their performance in adversarial settings remain significant unsolved challenges. I am enthusiastic about delving into these phenomena, specifically focusing on the over-sensitivity of such models to adversarial examples and their under-sensitivity to blind spots. I am also interested in developing efficient attacks and robust training techniques that have a relatively small computational footprint.

Before joining UMD as a PhD student, I earned both my Bachelor’s and Master’s degrees in Mathematics at The Indian Institute of Science (IISc), Bangalore. I was fortunate to be advised by Prof. Venkatesh Babu at the Computational and Data Sciences Department, and by Prof. Kaushal Verma at the Mathematics Department during my studies at IISc.