I joined Mila and the Computer Science and Operations Research department (DIRO) at the Université de Montréal (UdeM) as a research Master’s student in Sep. 2020. I’m advised by Irina Rish.
During my master’s I mostly focused on causal representation learning and structure learning and how they influence one another. I have studied factor graphs and how we could incorporate the inductive bias of reusability into them. As to causal representation learning, I have studied structured representations such as set-based representations in visual scenes, and how we can disentangle their properties. I have also worked on structured latent representations in the domain of language by studying structured text data in autoencoder architectures. For a list of publications, please visit here.
I spent the summer of 2021 at the data science lab at EPFL joining the Summer@EPFL program, studying discrete and sequential bottleneck structures with structured text to learn a Facts-as-Experts model.
Prior to Joining Mila, I did my bachelor’s in Electrical Engineering, Digital Systems Specialization at Sharif University of Technology in Tehran. During my undergrad, I was lucky to participate in the research of leading institutes in Switzerland and Japan. In the summer of 2019, I joined Riken AIP in Fukuoka, Japan, to work on a novel method for boosting based on zero-suppressed decision diagrams. I was co-supervised by Kohei Hatano and Eiji Takimoto. In the summer of 2018, I joined the amazing institute of Neuroinformatics at ETH Zurich and the University of Zurich. I was supervised by Yulia Sandamirskaya and her Ph.D. student, Raphaela Kreiser. We designed a Spiking Neural Network (SNN) inspired by known mechanisms in rodents to be deployed on neuromorphic hardware as a part of an agent’s localization and mapping system.
Currently, I’m interested in causal representation learning and structure learning.
Apart from AI research, find some of my piano playing recordings here!