Final project for the graduate course
CSC2541 - Topics in Machine Learning: Neural Network Training Dynamics by Roger Grosse
Taken as an elective for the degree of Master of Science, Department of Medical Biophysics, University of Toronto. Please note that this project was done using the ML-Commons algorithmic efficiency framework as part of a contribution to their codebase. Additionally two other groups also contributed within the same forked repository. Since it is difficult to differentiate our contributions from the much larger overall codebase, we seperately provide the final paper here for clarity. For those interested in the implementation, the working repository for this project can be found here. Thanks to the team at Google Brain for advising the project's direction and helping to onboard us with the ML-Commons algorithmic efficiency framework. Also thankyou to Dr. Roger Grosse for his feedback throughout the project.