I recently finished my third year of undergrad at University of Toronto and currently on a year of machine learning research internship at Vector Institute and Borealis AI where I am extremely fortunate to work with Roger Grosse, Sageev Oore, Alireza Makhzani and Yanshuai Cao. I spent my first two years at U of T in Engineering Science and after that I switched to study computer science, cognitive science, statistics and mathematics and started to work on machine learning research projects with Vector and Borealis. I'm interested in anything related to machine leanring, but currently mainly dabbling in information theory, generative models, network compression, reinforcement learning, optimization and cognitive science. In my free time, I'm excited about sharing my thoughts and knowledge and discussing interesting ideas. I started a machine learning student organization called UTMIST and I'm involved with FOR.ai, and I also TA'ed undergrad calculus and linear algebra where I tried to explain backprop in my tutorial to first year students. If you want to chat about anything with me, feel free to shoot me an email through the contact me section.
"Today, a group of University of Toronto undergraduate computer science students are decoding encrypted text using a neural network, a framework for machine learning algorithms inspired by the brain...." Read more
This work details CipherGAN, an architecture inspired by CycleGAN used for inferring the underlying cipher mapping given banks of unpaired ciphertext and plaintext. We demonstrate that CipherGAN is capable of cracking language data enciphered using shift and Vigenere ciphers to a high degree of fidelity and for vocabularies much larger than previously achieved. We prove that the technique used in CipherGAN avoids the common problem of uninformative discrimination associated with GANs applied to discrete data.Published as a conference paper in ICLR2018
Working with Yanshuai Cao on machine learning research projects.
Held a Teaching Assistantship in a course in calculus(MAT135) during the summer of my second year and a teaching assistantship for a course in linear algebra(MAT223) during 2017 fall.
Worked with Professor Qian Li on a research project of mathematical modelling of Brillouin amplification in elliptical core spun fiber at the Ultrafast Photonics Laboratory at the University of Toronto.