I had an opportunity to go through the theories behind many machine learning techniques from the course I took this semester, ISYE 6740 Computational Data Analysis from Prof Kai Wang. There I learned that many machine learning algorithms are basically a mixture of applied linear algebra and nonlinear optimization problems; each data point are represented as a vector and we have to minimize some loss functions (or maximize likelihood functions) expressed in matrix multiplications.
Thereby, I realized it is crucial to equip oneself with computational methods of linear algebra to be able to freely use machine learning algorithms. Recently, in an effort to do so, I picked up this lecture called Linear Algebra with Python - Using NumPy and SciPy to learn python libraries such as numpy and scipy for the relevant techniques. The lecturer is Bumhee Cho, who received a doctorate degree in nuclear engieering from KAIST and he seems to be working in the reactor core design department at KEPCO Nuclear Fuel.
Each jupyter notebook file contains python codes and relevant theorems in linear algebra.
Dec 31, 2024. (edited on Jan 31, 2025)