Hi, I am Shoaib!

---

I am an incoming Ph.D. student at the University of Cambridge supervised by David Krueger. Prior to this, I completed my BS in Computer Science from NUST (Islamabad, Pakistan) and MS in Computer Science from TU Kaiserslautern (Germany) while simultaneously working as a junior researcher at the German Research Center for Artificial Intelligence (DFKI). I also interned at NVIDIA Research (Santa Clara, USA) with Thomas Breuel working on adversarial robustness.          

I am broadly interested in understanding human intelligence, while simultaneously replicating intelligent behavior in machines to solve real-world problems. Human-inspired systems can more naturally incorporate human preferences and values (this can potentially be a solution to the model alignment problem as most of these simple alignment problems arise due to drastic differences in terms of processing between human brains and artificially intelligent systems). My core research interests lie in the development, understanding, and application of deep learning models. I am currently exploring different aspects of deep models including their robustness (against both natural as well as adversarial image degradations), interpretability, and representability (including the relation between the representation quality and their capacity, their memorization capabilities, as well as learning of these representations in a self-supervised way).          

An up-to-date list of my publications can be found through my Google scholar profile. A copy of my CV can be found here. I am always happy to work on interesting research problems together with others. Therefore, if you are interested in collaborating, please feel free to reach out through my email address.