Skip to content

Latest commit

 

History

History
65 lines (49 loc) · 1.98 KB

README.md

File metadata and controls

65 lines (49 loc) · 1.98 KB
Raw

Machine Learning Course - Spring 1403

This repository contains all the materials, projects, and assignments for the Machine Learning course taught by Dr. Aliyari at the Faculty of Electrical and Computer Engineering, K. N. Toosi University of Technology, during Spring 1403.

Course Topics

The following topics were covered in this course:

Optimization

  • Batch, Mini-Batch, and Online updates
  • Gradient Descent
  • Newton Method & Hessian Matrix
  • Levenberg
  • Momentum
  • PSO
  • Adam

Regression & Classification

  • Linear Regression
  • Log Likelihood
  • Bayes Classifier

Neural Networks

  • Different kinds of activation functions
  • Batch Normalization
  • Overfitting & Overmodeling
  • Dropout
  • Different NN models like MLP, RBF

Data Preparation & Validation

  • Data Splitting
  • K-fold Cross Validation
  • Unbalanced Data & Bouts Trapping

Decision Trees & Ensemble Methods

  • Decision Trees
  • Random Forest
  • AdaBoost

Support Vector Machines

  • SVM & Vapnik with soft or hard margin
  • Kernel Tricks & Mercer Theorem

Feature Engineering

  • Feature Extraction
  • Feature Selection (Forward Selection, Backward Elimination, Fisher)
  • Non-linear Feature Mapping (PCA, LDA, AE)

Reinforcement Learning

  • Introduction to Reinforcement Learning

Repository Structure

The repository is organized into the following branches:

  • mini-project-1: Contains materials and code for the first mini project.
  • mini-project-2: Contains materials and code for the second mini project.
  • mini-project-3: Contains materials and code for the third mini project.
  • mini-project-4: Contains materials and code for the fourth mini project.
  • final-project: Contains materials and code for the final project.

Note! : All the codes for this course are implemented in Python in the Google Colab environment. The code files for each mini project are placed in their respective branches.

Content Information

For any questions or further information, please contact [email protected]