CSAL4243 Introduction to Machine Learning
These notes accompany the University of Central Punjab CS class CSAL4243: Introduction to Machine Learning. The course provides an introduction to machine learning i.e. how to make computers learn from data without being explicitly programmed. Topics include: supervised learning (regression and classification, parametric/non-parametric learning, neural networks, and support vector machines); unsupervised learning (clustering, dimensionality reduction); model optimization and and best practices (bias/variance tradeoffs);. The course will also discuss advanced topics in machine learning, like deep learning, convolutional neural networks and large scale machine learning.
Reference Material
Sebastian Raschka, Python Machine Learning. 2015
James, G., Witten, D., Hastie, T., Tibshirani, R., An Introduction to Statistical Learning with Applications in R. Springer, 2013
Trevor Hastie, Robert Tibshirani and Jerome Friedman, The Elements of Statistical Learning. Springer, 2009
Kevin P. Murphy, machine learning a probabilistic perspective. MIT press, 2012
Andrew Ng, Machine Learning, Coursera
Sebastian Thrun, Intro to Machine Learning, Udacity
Tutorials
Python / Numpy Tutorial
IPython Notebook Tutorial
Python Setup
Anaconda Environment in Windows
Lectures
  1. Introduction to Machine Learning [Notes] [Video] [Notebook]
  2. Regression with One Variable [Notes] [Video] [Notebook]
  3. Gradient Descent [Notes] [Video] [Notebook]
  4. Linear Regression Example [Notes] [Video] [Notebook]
    Quiz 1a
    Quiz 1b
  5. Multivariate Regression [Notes] [Video] [Notebook]
  6. Classification and kNN [Notes] [Video] [Notebook]
  7. Logistic Regression [Notes] [Video]
  8. Logistic Regression Cost Function [Notes] [Video] [Example]
  10. Data Preprocessing [Notes] [Video] [Notebook]
  11. Neural Networks - Introduction [Notes] [Video]
  12. Neural Networks - Forward Propagation [Notes] [Video] [Play with Neural Network]
  13. Midterm Exam
  14. Neural Networks - Back Propagation [Notes] [Video]
  15. Revision Class
    Quiz 3
  17. Convolutional Neural Network (CNN) [Notes] [Video] [How CNN Work]
  18. tflearn and Tensorflow [tflearn] [Video] [MNIST with CNN using tflearn]
    Quiz 4
  19. Support Vector Machines (SVM) - Introduction [Notes] [Video1] [Video2]
  20. Support Vector Machines (SVM) - Kernels [Notes] [Video1] [Video2]
  21. Project Progress Presentation
    Schedule
  22. Unsupervised Learning - Clustering [Notes] [Video1] [Video2]
  23. Ensemble Learning [tflearn] [Video] [Read chapter 7 of Sebastian Raschka's Introduction to Machine Learning]
    Quiz 4
  24. Dimensionality Reduction - PCA (Pricipal Component Analysis) [Notes] [Video1] [Video2]
Assignments
  1. Assignment 01: House price prediction using Linear Regression
  2. Assignment 02: Digit recognizer using Logistic Regression and Neural Network
  3. Assignment 03: Convolutional Neural Network for image classification on CIFAR-10
    [Download]
Projects
List of projects done by students as part of the course are given below. Github repository for each project can be reached by clicking on the project name.
  1. Vehicle detection using Convolutional Neural Network
  2. Identify signs of diabetic retinopathy in eye images
  3. Teaching a car to drive itself
  4. Distracted Driver Detection
  5. Lung cancer detection in images
  6. Human detection in video
  7. Face recognition system
  8. Vehicle classification in images
  9. Hand written Urdu characters recognition
  10. Student performance prediction
  11. House price prediction
  12. Diabetes prediction from patient vitals
  13. Why employees leave early
  14. Breast Cancer prediction as benign or malignant
  15. Face recognition system
  16. Traffic sign detection
  17. Facial Keypoint detection
  18. Spam email classifier using WEKA
Credits
Credit for the website design and outlook goes to Fei-Fei Li and Andrej Karpathy Stanford course cs231n: Convolutional Neural Networks for Visual Recognition