Course Projects
Clustering Oriented Representation Learning in Neural Networks. Final project for COMP 652: Machine Learning, taught by Guillaume Rabusseau. Obtained a grade of 95%. This represents my preliminary work on clustering oriented representation learning; while this paper is somewhat more poetically written than a typical ML paper, and although my understanding of this technology has advanced since its creation, it still may be an interesting read for some people. See the PDF here; github repository here.
Trouble in Gaussian City: When Laser-Tanks Use Potential Fields. Final project for COMP 521: Modern Computer Games, taught by Clark Verbrugge. Obtained a grade of 97%. Explored the use of potential field algorithms as a way to allow for super-rapid group path-finding algorithms (in this case, groups of enemy tanks that seek to hunt down the player). See the PDF of the report for more information.
Presentations
A Brief Survey on Word Embedding Methods Presentation given to CompLing lab at McGill on August 6th, 2018; here.
Sentiment Analysis: It's Complicated! Presentation on my NAACL 2018 paper at NAACL in New Orleans on June 4th, 2018; slides here.
Resolving Event Coreference with Supervised Representation Learning and Clustering-Oriented Regularization. Presentation on my *SEM paper given in New Orleans at *SEM on June 5th, 2018; slides here.
Clustering Oriented Representation Learning in Neural Networks. Presentation given to the McGill RLLab; slides here.
Miscellaneous
Some notes I took while reviewing for COMP 767 Reinforcement Learning exam, certainly no guarantees that these are completely correct, robust, or readable.
Some notes I took while brushing up on linear algebra, no guarantees that these old notes are completely correct, robust, or readable.
Some notes I took while brushing up on advanced calculus, no guarantees that these old notes are completely correct, robust, or readable.