Hi! I’m Mohammad, a Ph.D. student at the University of Amsterdam working on problems related to artificial intelligence under the supervision of Prof. Cees Snoek and Dr. Xiantong Zhen. My Ph.D. thesis is Continual Learning, in which I attempt to train neural networks to tackle a sequence of non-stationary problems. Since last two years, I have been studying continual learning from the viewpoints of probabilistic modelling (variational inference), non-parametric approaches (kernel methods), and generative models (variational auto-encoder and energy-based models).
Currently, I am an ML/CV intern at Samsung AI Cambridge under the supervision of Dr. Brais Martinez and Dr. Georgios Tzimiropoulos doing research on Federated Learning and the prompting of large-scale language-image models. Specifically, I intend to identify the most informative and effective prompt for few-shot learning, cross-domain generalisation, and robustness. I have also been recently intrigued by the performance of large-scale language-image models such as CLIP and also the diffusion model, particularly their prompts and in-context learning capabilities.
I am also a member of ELIS society. Previously I was a master student at the University of Tehran working with Prof. Babak Nadjar Araabi and Dr. Mohammad Amin Sadeghi. I did my master at machine learning and computational modeling lab working on object detection and image compression. During my master degree, I also did research on object detection with Dr. Mohammad Rastegari at Allen AI Institute.