Master's Machine Learning
Faculty
Sergey Nikolenko
Chief Research Officer, Neuromation Head of AI Lab, PDMI RAS
Course length
Duration
Total hours
Credits
Language
Course type
Fee for single course
Fee for degree students
Skills you’ll learn
Overview
Machine learning is one of the most rapidly developing, most useful for applications, and generally “hottest” fields of computer science. However, it is not simply a collection of ad hoc recipes but a precise science based on probability theory. In the course, we will look at some of the most widely used machine learning tools through a probabilistic lens, understanding the intuition behind them, their applicability, and how to choose between them and apply them in real world situations. The course is accompanied with practical data science, where we will train the models on practical datasets, using both standard libraries provided in the Python ecosystem and other tools.
Learning highlights
- Become well versed in modern probabilistic inference, a foundation of machine learning.
- Learn basic and generalized linear models, SVMs, models for unsupervised learning (clustering and HMMs), probabilistic graphical models and approximate inference for PGMs, basic reinforcement learning, and several case studies covering some application areas that give rise to other interesting models and algorithms.
- Understand the probabilistic intuition behind all of these models, learn to apply them correctly.
- Learn to apply these skills to practical settings, analyzing real life datasets.learn to apply these skills to practical settings, analyzing real life datasets.
Course outline
15 classes
Intro and Bayes theorem
Introduction. History of AI. Types of machine learning problems. Probability theory basics. Bayes’; theorem and maximal a posteriori hypotheses. Laplace’s rule of succession.
Linear Regression
Gaussian distribution, its ML estimates. The multidimensional Gaussian. Linear regression and least squares. Least squares as an ML estimate for Gaussian noise.
Intro and Bayes theorem
Overfitting. Regularization. Ridge regression and lasso regression. Bayesian view of linear regression: MAP and predictive distribution.
Classification
Classification: 1-of-K representation, linear decision functions. Fischer's linear discriminant. Bayes theorem for classification. Logistic regression. Multiclass logistic regression and softmax. The Bayesian view of logistic regression.
Support vector machines
Support vector machines. Linear separation and max-margin classifiers. Quadratic optimization. Kernel trick. SVM variations: ν-SVM, one-class SVM, SVM for regression.
Statistical decision theory
Regression function, optimal Bayesian classifier. Bias-variance-noise decomposition. Nearest neighbors and the curse of dimensionality
Model selection
Model selection via Laplace approximations.
Bayesian information criterion. Examples.
Model combination
How to construct ensembles of models. Bayesian model averaging. Bootstrapping and bagging. Boosting: AdaBoost, gradient boosting.
Unsupervised learning
Clustering. The EM algorithm for clustering. Justification of the EM algorithm Hidden Markov models and the Baum-Welch algorithm.
Probabilistic graphical models
Probabilistic graphical models: basic idea, factorizations, d-separation. Directed and undirected models. Factor graphs. Inference on factor graphs. Belief propagation with message passing.
Model selection
Model selection via Laplace approximations.
Bayesian information criterion. Examples.
Model combination
How to construct ensembles of models. Bayesian model averaging. Bootstrapping and bagging. Boosting: AdaBoost, gradient boosting.
Unsupervised learning
Clustering. The EM algorithm for clustering. Justification of the EM algorithm Hidden Markov models and the Baum-Welch algorithm.
Probabilistic graphical models
Probabilistic graphical models: basic idea, factorizations, d-separation. Directed and undirected models. Factor graphs. Inference on factor graphs. Belief propagation with message passing.
Theme: Reinforcement learning
Multiarmed bandits, exploration vs. exploitation. Confidence intervals. Minimizing regret: UCB1. Markov decision processes. On-policy and off-policy learning. TD-learning.
Prerequisites
This course is one of three in a wholistic series.
Students that have already taken MSL-111 and those with prior experience with HTML, CSS, and Javascript building simple web pages will be good candidates for this module.
Sergey Nikolenko is a computer scientist with vast experience in machine learning and data analysis, algorithms design and analysis, theoretical computer science, and algebra. He graduated from St. Petersburg State University in 2005, majoring in algebra (Chevalley groups), and earned his Ph.D at the Steklov Mathematical Institute at St. Petersburg in 2009 in theoretical computer science (circuit complexity and theoretical cryptography). Since then, Sergey has been interested in machine learning and probabilistic modeling, producing theoretical results and working on practical projects for the industry.
Sergey Nikolenko is currently serving as the Chief Research Officer at Neuromation, leading the Artificial Intelligence Lab at the Steklov Mathematical Institute at St. Petersburg, and teaching at the St. Petersburg State University and Higher School of Economics. Dr. Nikolenko has published more than 170 research papers on machine learning (ICML, CVPR, ACL, SIGIR, WSDM...), analysis of algorithms (SIGCOMM, INFOCOM, ICNP…), and other fields, several books, including a bestselling “Deep Learning” book (in Russian), lecture courses in ML, DL, other fields of computer science (St. Petersburg State University, NRU Higher School of Economics...) and much more. He has extensive experience in managing research and industrial AI/ML projects.
See full profileApply for this course
Master's Machine Learning
by Sergey Nikolenko
Total hours
45 Hours
Dates
Mar 13 - Mar 31, 2017
Fee for single course
€1500
Fee for degree students
€750
How to secure your spot
Complete the form below to kickstart your application
Schedule your Harbour.Space interview
If successful, get ready to join us on campus
FAQ
Will I receive a certificate after completion?
Yes. Upon completion of the course, you will receive a certificate signed by the director of the program your course belonged to.
Do I need a visa?
This depends on your case. Please check with the Spanish or Thai consulate in your country of residence about visa requirements. We will do our part to provide you with the necessary documents, such as the Certificate of Enrollment.
Can I get a discount?
Yes. The easiest way to enroll in a course at a discounted price is to register for multiple courses. Registering for multiple courses will reduce the cost per individual course. Please ask the Admissions Office for more information about the other kinds of discounts we offer and what you can do to receive one.