Machine Learning

This is the repository for the course Machine Learning at Asian Institute of Technology.

Prerequisites

• Visit our "Prerequisites" folder to review the materials, before attempting our ML course

Technology stack

• NumPy, Pandas, Matplotlib, Sklearn, PyTorch - for machine and deep learning
• MLFlow - for experimenting
• FastAPI - for exposing the models
• Anything for frontend, e.g., Vue, ReAct, Angular, Jinja, Hugo, etc.
• Anything for backend, e.g., Django, Flask
• Docker for containerization, and Traefik for reverse proxy

Outline

The course is structured into 5 big components:

0. Case Study

• Regression
• Classification

1. Supervised Learning

Regression

• Gradient Descent
• Stochastic and Mini-batch
• Regularization

Classification

• Logistic Regression
• Naive Bayes
• K-Nearest Neighbors
• Support Vector Machines
• Decision Trees
• Random Forest
• AdaBoost
• Gradient Boosting

2. Unsupervised Learning

• K-mean clustering
• Gaussian mixture
• Principle component analysis

3. Deep Learning

• Feedforward Neural Netork
• Convolutional Neural Network
• Recurrent Neural Network

4. Reinforcement Learning

• PPO

References:

• Hastie, T., Tibshirani, R., and Friedman, J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2016 (2nd edition) (https://web.stanford.edu/~hastie/)
• Ian Goodfellow., Deep Learning, 2016
• Pytorch tutorials available online: https://pytorch.org/tutorials/
• Nice visuals on CNN - https://github.com/vdumoulin/conv_arithmetic