Machine Learning - Theory and Practice
This is the course web page for the ACA Summer school module : Machine Learning Theory and Practice. Please follow the links you can see around you to navigate.
Announcements and links
Post course feedback : Please fill this extremely short survey if you were part of my class. It would help in improving the offering of this course in the future, as well as provide me with feedback.
Programming tutorials for (PCA, SVM with Kernels, Adaboost) are all up. Please find them from the code tab to your top right.
Some students have mentioned the difficulty in downloading Anaconda. I have hosted a local copy here. Please install the relevant package for your architecture.
Programming Assignment 2 is up. Please find it in the “Code” tabe to the top right. It is meant as an introduction to gradient descent and loss functions. Please let me know if there are any errors in the code, since I can’t claim to be an expert at coding.
Quiz 1 is up. Please find it here. It consists of a few short answer questions, mostly the true or false kind. It is automatically graded, and should also give you hints as to why your answer is wrong. Please let me know if you have any difficulties with the quiz.
Programming Assignment 1 is up. Please find it in the “Code” tabe to the top right. It is meant only as an introduction to scikit-learn and the IPython notebook environment.
Pre-course survey - Please fill up this form, it lets me send you course emails directly, and also adjust the course contents according to student demographic.
Permanent Feedback - Please let me know of any complaints / feedback you may have about the course (method of delivery, matter delivered). You can choose to identify yourself if you wish.
Tentative Schedule :
- Introduction to the course - Logistics, Background
- Machine Learning Basics - Gradients, Inversions, Similarities
- Supervised learning - KNN, Decision Trees
- Supervised learning - Linear Regression (and Matrix Factorization)
- Supervised learning - Logistic regression
- Unsupervised learning - KMeans, GMMs
- Unsupervised learning - PCA, Kernels, Ensembles
- Advanced methods - SVM’s, Neural Networks
- Feature engineering - Dealing with Images, Text, Video
- Practical ML - When and why 100% accuracy is bad