From the course outline:
This course therefore aims to provide a gentle introduction to foundational concepts in optimization with a strong emphasis on insights and applications to problems in data science and machine learning. The course spans fundamental optimization concepts and models such as likelihood functions, loss functions, linear, integer, and convex programming, a variety of gradient descent methods, and heuristic algorithms such as the E-M algorithm.
Instructor: Prof. Sanjit Gopalakrishnan
This repository contains my solutions to the assignments for the course MGSC 695: Optimization for Data Science at McGill University. The assignments are an implementation of machine learning methods from scratch, using Python and NumPy, and comparing them to the performance of the same methods from the scikit-learn library.