SDS 404: Machine Learning and its Applications
Course Title |
Machine Learning and its Applications |
||||||
Course Code |
SDS 404 |
||||||
Course Type |
Mandatory |
||||||
Level |
Master’s |
||||||
Year / Semester |
1st / 2nd (subject to change) |
||||||
Instructor’s Name |
Mihalis Nicolaou |
||||||
ECTS |
10 |
Lectures / week |
1 (2h) |
Laboratories / week |
1 (2h) |
||
Course Purpose and Objectives |
The aim of this course is to provide a broad introduction to students on both theoretical as well as practical concepts in machine learning, data mining and pattern recognition. Topics include fundamental machine learning concepts and algorithms, such as supervised learning (parametric and non-parametric algorithms, classification and regression, discriminative and generative learning), unsupervised learning (clustering, dimensionality reduction, data imputation), and learning theory (bias-variance tradeoff, curse of dimensionality). The course will also include an introduction to deep learning, practical advice for designing machine learning systems, as well as an overview of modern scientific applications of machine learning and data mining (e.g., classification of omics data and applications in biology, object detection and human behaviour analysis, weather forecasting). |
||||||
Learning Outcomes |
By the end of the course, students will be able to demonstrate a critical understanding of fundamental concepts in machine learning and data mining, as well as gain practical skills in applications related to a variety of scientific application domains. Students will be familiarized with a set of core algorithms to machine learning and data mining, be able to select, implement and apply the appropriate algorithms based on problems, applications and datasets, and be familiarized with extracting appropriate feature representations and rigorous evaluation of algorithm performance. |
||||||
Prerequisites |
SDS 401, SDS 402, |
Requirements | None | ||||
Course Content |
Introduction to Machine Learning and Data Mining: Supervised learning (parametric and non-parametric algorithms, classification and regression, discriminative and generative learning), unsupervised learning (clustering, dimensionality reduction, density estimation, data imputation), learning theory (bias-variance tradeoff, curse of dimensionality). Exploration of linear and non-linear learning (kernel methods, neural networks). Introduction to reinforcement learning. Introduction to deep learning (convolutional, recurrent networks, adversarial learning). Machine Learning Applications: Practical advice for designing machine learning systems (e.g., for big data, combining heterogeneous data sources, on-line learning). Students will carry out practical data-driven projects, utilizing data science tools in scientific applications such as biology (classification of omics data and sequence analysis), computer vision (human sensing, facial image analysis, object detection), physics (Ising model phase transitions), energy (solar forecasting) and weather modelling (daily rainfall, weather forecasting). |
||||||
Teaching Methodology |
Lectures, exercises |
||||||
Bibliography |
C. Bishop, "Pattern Recognition and Machine Learning", ISBN 978-0-387-31073-2, 2009 K. Murphy "Machine Learning: a Probabilistic Perspective", MIT Press, 2012 I. Goodfellow, Y. Bengio, A. Courville “Deep Learning”, MIT Press, 2017 G. James, D. Witten, T. Hastie and R. Tibshirani, “An Introduction to Statistical Learning”, ISBN-13: 978-1461471370, 2017 I. Witten, E. Frank, M. Hall, “Data Mining: Practical Machine Learning Tools and Techniques”, ISBN-13: 978-0123748560, 1999 |
||||||
Assessment |
25% coursework and 75% exam |
||||||
Language |
English |