Y. Emre Yildiz

syllabus - Intro. to Machine learning

Course Description. Machine Learning (ML) is the design of a system that can learn from data. This course covers the basics of ML such as Supervised Learning, Unsupervised Learning, and Reinforcement Learning.
Coursework. Coursework will consist of weekly homework, quizzes , midterm, and a final exam. The overall grade will be determined %10 from class activity, 10% for each quiz, 10% from midterm, and 60% from the final exam.
Exam policy. No collaboration is permitted during the exam . If any collaboration with the intention of copying is caught, the student will get a failing grade.
Smartphone policy. Smartphones are not allowed during lectures.
Text. The course textbook is "Introduction to Machine Learning", Third Edition, by Ethem Alpaydin. As an additional text book: "Introduction to Machine Learning with Python" by Andreas C. Mueller and Sarah Guido,

Week 1 - Introduction
1. Why machine learning?
2. A First Application: Classifying iris species
Week 2 - Supervised Learning
1. Classification and Regression
2. Generalization, Overfitting and Underfitting
3. Supervised Machine Learning Algorithms
4. k-Nearest Neighbor
5. Linear models
Week 3 - Supervised Learning
1. Naive Bayes Classifiers and Gaussian class-conditional distribution
2. Decision trees and Ensembles of Decision Trees
3. Kernelized Support Vector Machines
Week 4 - Supervised Learning
1. Logistic regression, gradient descent, Neural Networks (Deep Learning)
2. Uncertainty estimates from classifiers
3. Quiz 1
Week 5 - Unsupervised Learning and Preprocessing
1. Types of unsupervised learning
2. Preprocessing and Scaling
Week 6 - Unsupervised Learning and Preprocessing
1. Dimensionality Reduction, Feature Extraction and Manifold Learning
2. Advanced discussion on clustering and EM
Week 7 - Representing Data and Engineering Features
1. Categorical Variables
2. Binning, Discretization, Linear Models and Trees
3. Interactions and Polynomials
4. Midterm
Week 8 - Representing Data and Engineering Features
1. Univariate Non-linear transformations
2. Automatic Feature Selection
3. Utilizing Expert Knowledge
Week 9 - Model evaluation and improvement
1. Cross-validation
2. Grid Search
Week 10 - Model evaluation and improvement
1. Evaluation Metrics and scoring
2. Using evaluation metrics in model selection
Week 11 - Algorithm Chains and Pipelines
1. Parameter Selection with Preprocessing
Week 12 - Working with Text Data
1. Types of data represented as strings
2. Rescaling the data with TFIDF
3. Topic Modeling and Document Clustering
4. Quiz 2
Reinforcement Learning
Reinforcement Learning

Google Sites

Report abuse