riquitorres / machine-learning-methods

Implementations (from scratch), applications, and visualizations of key machine learning methods from UBC CPSC 532M/340: Machine Learning & Data Mining and UBC CPSC 540/440: Advanced Machine Learning

Machine Learning Implementation & Applications

A collection of from-scratch implementations of traditional ML models, insightful visualizations, and mathematical derivations completed independently during both undergraduate ML courses at UBC Vancouver:

• UBC CPSC 532M/340: Machine Learning and Data Mining (Jan - Apr 2022)
• UBC CPSC 540/440: Advanced Machine Learning (Jan - Apr 2023)

Simple Regression

Linear Regression, Robust Linear Regression, Polynomial Bases

Further Regression

Logistic Regression, Softmax Classifiers, Convexity

Advanced Unsupervised Learning

Robust PCA, Collaborative Filtering

Miscellaneous

PCA, MAP Estimation, Stochastic Gradient Descent

Markov Chains & Monte Carlo Methods

Ancestral sampling, Marginal and Conditional Probabilities, Viterbi Decoding, Monte Carlo Approximation

Generative Classifiers

Gaussian Discriminant Analysis, Discriminant Analysis with Student-t

Bayesian Methods

Vector-quantized Naive Bayes, Posterior-predictive Probability

Deep Learning

Neural networks, 10-way MNIST Classification, ConvNet