Courses

CS577 Machine Learning

Title: Machine Learning
Course Number: CS577
Level: Graduate

Description: Machine Learning is a vibrant area of Computer Science with thousands of applications to real problems, ranging from predicting the stock-market, diagnosing disease, teaching autonomous helicopters to fly, and understanding biological mechanisms. The goal of the course is to introduce the theory, main principles, methods, and algorithms of Machine Learning, but also the tools and practical aspects of data analysis. The students become familiar with the subject by a series of practical assignments, theory exercises, and a course project. The main content of the course is:

1. Supervised learning and learning through examples: algorithms for learning classification models, including the Simple Bayes Classifier, Decision Trees, K-Nearest Neighbors, Artificial Neural Networks, and Support Vector Machines.
2. Methods for measuring performance, particularly the Area Under the Receiving Operating Characteristic Curve, model selection and parameters optimization, as well as estimation of model performance, using cross-validation and nested cross-validation techniques.
3. Causal Discovery based on Bayesian Networks and Maximal Ancestral Graphs, and single statistical hypothesis testing.
4. Other subjects depending on time availability such as variable selection, unsupervised learning, clustering, and others.
5. Exposition to other types of learning problems, areas, techniques and approaches, such as Reinforcement learning, and relational learning

Instructor: Ioannis Tsamardinos
Link: http://elearn.uoc.gr/course/view.php?id=321

CS387 Introduction to Artificial Intelligence

Title: Introduction to Artificial Intelligence
Course Number: CS387
Level: Junior (third year)

Description: Artificial Intelligence (AI) refers to a corpus of techniques developed to simulate or emulate natural intelligence or to enhance reasoning capabilities in software agents. AI is a vibrant area of Computer Science with thousands of applications to designing intelligent systems or to solving problems in other areas of Computer Science. The goal of the course is to introduce the theory, main principles, methods, and algorithms of AI, but also to present some of the tools and practical aspects of the application of AI algorithms and techniques. The students become familiar with the subject by a series of practical assignments and theory exercises. The main content of the course is:

1. Problem solving using search: uninformed search, informed search and constraint-satisfaction techniques
2. Propositional and First Order Logic based agents and reasoning
3. Planning
4. Introduction to Probabilistic reasoning and Decision Making for agents.

Instructor: Ioannis Tsamardinos
Link: http://elearn.uoc.gr/course/view.php?id=211

CS482 Algorithms in Bioinformatics

Title: Algorithms in Bioinformatics
Course Number: CS482
Level: Senior (fourth year)

Description: Bioinformatics is the branch of Computer Science that applies computational methods to enable, improve, complement, and facilitate biological research. Modern biology, and particularly molecular biology, is virtually impossible without the use of computational methods. Bioinformatics, in conjunction with progress in biology and bioengineering, is continuously our shaping our society through new discovering. The goal of the course is to introduce provide an introduction to the subject and present some of the basic algorithms in the field. The students become familiar with the subject by a series of practical assignments and theory exercises. The main content of the course is:

1. Dynamic Programming algorithms for single and multiple sequence alignments
2. Graph algorithms in bioinformatics for optimization and visualization of biomedical networks (such metabolic networks, gene interaction networks, evolutionary trees)
3. Single and multiple hypothesis testing for identifying differentially expressed genes
4. Modern and basic clustering algorithms and applications to the analysis of biological data
5. Hidden Markov Models
6. Invited lecturers will present their state-of-the-art research

Instructors: Ioannis Tsamardinos and Ioannis Tollis
Link: http://elearn.uoc.gr/course/view.php?id=280