Technologies of Industrial Software Development and Production

Entry requirements: Basic modelling and programming skills

Credits: 4

Course: Core

Language of the course: Russian

Objectives

Students will learn:

  • To perform design supervision of engineering, implementation and maintenance of information systems and technologies
  • To implement new principles of building infocommunication systems and networks of various types of data transmission, distribution, processing and storage
  • To perform research, engineering, to organize production processes, to use infocommunication systems, networks and devices
  • To develop theoretical and experimental models for objects of professional activity in financial markets and financial technologies

Contents

  • Classifications of technologies of industrial software development and production: Introduction to machine learning. Bias-Variance dilemma. Ridge regression, lasso method, LARS. Logistic regression. Naive Bayes model. Metric methods of classification and regression. Nearest neighbours methods. Parzen windows methods. Logical classification algorithms. Support Vector Machines. Decision tree algorithms. Ensemble methods.
  • Methods of data clustering. k-means method. Expectation-Maximization algorithm. Introduction to association analysis. Apriori algorithm. FP-Growth algorithm. Dimensionality reduction methods.
  • Neural networks: Multilayer perceptron. Stochastic gradient method. Back propagation of error method. Optimal brain damage. Hopfield neural network. Restricted Boltzmann machines. Contrastive Divergence algorithm. Deep learning. Deep belief networks. Restricted Boltzmann machines for deep belief networks. Convolutional neural networks. Kohonen map. Recurrent neural networks.

Format

Lectures and labs

Assessment

Attendance is mandatory. Students should complete all the assignments. The final grade is based on the student performance throughout the course.