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