ҚазРусEng
ҚазРусEng

Computational Science

7M06104 Computational Science

Core subjects: algorithms and data structures, as well as databases.

The graduate program «7M06104 Computational Science» is designed to provide students with a comprehensive understanding of the interdisciplinary field that combines computer science, mathematics, and scientific research. This program equips students with the knowledge and skills to develop computational models and simulations to solve complex scientific and engineering problems. Through coursework and hands-on projects, students gain expertise in programming, algorithm development, data analysis, and high-performance computing. Graduates of this program are prepared for careers in fields such as scientific research, engineering, data analysis, and software development, where computational approaches play a crucial role in advancing knowledge and solving real-world challenges.

Contacts

Admission Committee

(7172) 64-57-10
info@astanait.edu.kz

Mon-Fri 9:00 – 18:00

Objective of Educational Program

The goal of the educational program is to provide training of highly qualified scientific and applied specialists and software engineers in the field of modeling, algorithms and analysis of large data, as well as managers and managers of software and information systems for the information technology industry and interdisciplinary industries related to the protection and processing of data in various sectors of the economy of the Republic of Kazakhstan.

List of a specialist’s positions

Career opportunities
  • Database Administrator;
  • Software Developer;
  • High Load Application Developer;
  • AI Developer;
  • HPC calculator;
  • Developer of Mathematical Models;
  • Computational Experiment Programmer;
  • Head of the organization;
  • Deputy head of the structural unit;
  • Head of the structural unit;
  • Expert of the republican center;
  • employee of the national scientific and practical center, university;
  • Engineer-developer of artificial neural networks;
  • Quantum Computing Technologist;
  • Engineer-technologist of peripheral computing;
  • Quantum Computing Analyst.

M094 – Information technology

Group of educational programs

Master of Technical Science in Educational Program "7M06104 Computational Science"

Awarded degree

2 years

Duration of study

Learning outcomes

  • Developing methods and algorithms for computational mathematics based on the approximation of differential equations by methods of finite differences, volumes or elements.
  • Conducting a fundamental analysis of computational methods and difference schemes for convergence and correctness, including in the case of high-performance algorithms using elements of mathematical logic and the theory of computability.
  • Solving computational problems with complex geometry of regions by building and using correct structured, curvilinear, unstructured computational grids.
  • Using data mining techniques based on deep learning, reinforcement learning, generative adversarial networks to effectively predict outcomes.
  • Developing parallel computational algorithms for engineering problems and implementing them in high-performance systems, developing quantum computing algorithms.
  • Developing and conducting computational simulations of probabilistic processes from various industries using stochastic modeling methods.
  • Using methods of data analysis in various areas of production, on real data for the selection of parameters, adaptation and testing of computing systems based on real experiments.
  • Conducting independent research, solving modern urgent problems, publishing results in high-rating journals and speaking at conferences.
  • Have knowledge in related fields of project management, data science, information security.

Documents

EDUCATIONAL PROGRAM DOCUMENTS

DEVELOPMENT PLAN

Academic disciplines

Cycle of fundamental disciplines

University’s component

Psychology of Management

The content of the course is aimed at the formation of systemic ideas about the psychological laws of management, the specifics of using social and psychological knowledge and mastering the skills of analyzing the social and psychological principles that underlie effective management.

Higher Education Pedagogy

The content of the course is aimed at acquiring knowledge about the foundations of pedagogical theory and pedagogical skills, the management of the educational process for teaching in higher education, the formation of an understanding of the main categories of pedagogy, the place, role and significance of higher education pedagogy, understanding the basic principles of modern pedagogy and methodological approaches to solving pedagogical problems of higher education.

Foreign language (professional)

The aim of the course is to develop professional competences of specialists; to form professionally oriented communicative competence of master’s students, which allows them to integrate into international professional environment and use professional English as a means of intercultural and professional communication.

Teaching Practicum (Internship)

Teaching practice is a kind of practical activities of graduate students, including teaching, organization of educational activity of students, scientific and methodical work on the subject, obtaining skills in teacher’s work.

History and Philosophy of Science

The content of the course is aimed at obtaining knowledge about the properties of science as a type of cognition and a socio-cultural phenomenon in its historical development by a master student; formation of system ideas about the general laws of scientific knowledge in its historical development and changing socio-cultural context.

Cycle of fundamental disciplines

Elective component

Introduction to neural networks

This discipline involves the study of the sections of mathematics and computer science necessary to introduce to the theory of machine learning and its section the theory of deep learning based on the backpropagation algorithm, which allows the AI system to learn itself. This section contains the tasks of image recognition, image and 3D generation , text , sounds, recognition and etc., it is one of the most common areas in modern machine learning.

Computational Geometry

This course introduces students to the concepts and techniques of computational geometry, including the representation, manipulation, and analysis of geometric objects. Students will learn about various algorithms and data structures for solving geometric problems, with applications in areas such as computer graphics, computer vision, robotics, and geographic information systems (GIS).

Partial Differential Equations

This course introduces students to the fundamentals of partial differential equations, including classifications, boundary and initial value problems, and analytical solution techniques. Students will learn how to apply PDEs to model and analyze a wide range of problems in areas such as fluid dynamics, heat transfer, and optimization.

Stochastic modeling

This discipline is directed to the basics of stochastic modeling, practical application of Monte Carlo methods, solving stochastic differential equations, probabilistic modeling for solving practical problems.

Markov chains and decision-making processes

This discipline involves the study of Markov chains, in which each element is completely determined by the previous one. These chains are widely used in the formulation of problems of linking artificial intelligence to the behavior of an agent in a certain environment, for example, a robot in a real environment, on which, for example, reinforcement learning is based. As a result of studying the discipline, the student must know: methods of constructing probabilistic models describing the stochastic dynamics of processes; queuing systems; be able to establish the properties of solutions to stochastic systems.

High-performance computing

The aim of the discipline is to study the fundamental techniques for developing HPC applications, the commonly used HPC platforms, the methods for measuring, assessing and analysing the performance of HPC applications, and the role of administration, workload and resource management in an HPC management software. The students will be introduced to the issues related to the use of HPC techniques in solving large scientific problems.

Cycle of major disciplines

University’s component

Mathematics for Computational Science

This discipline covers an introduction to mathematical courses necessary for mastering specialized disciplines of computational science based on numerical solutions of deterministic and probabilistic equations of mathematical physics and applied models used in technical production and the financial sector. Namely, it covers the theory of ordinary differential equations, their typification and basic methods of analytical solution and an introduction to partial differential equations.

Teaching methods and strategies

This course explores innovative pedagogical approaches and effective techniques for engaging and facilitating learning in diverse educational settings.

Numerical methods and computer modeling

This discipline involves the study of the fundamentals of numerical methods in the field of modeling physical processes, including algebraic numerical methods, numerical integration and numerical solution of partial differential equations, it also studies the introduction to methods of finite differences.

Applied mathematical models

This discipline focuses on common mathematical models used in manufacturing and their solution using numerical methods. Upon mastering the discipline, the student must know: basic mathematical models such as “predator-prey”, the equation of heat conduction, etc .; basic models of hydrodynamics, filtration, chemical reactions; be able to: approximate and investigate the convergence of the model; apply basic numerical methods to solve applied problems.

Research Practice

Research practice

Case study in computational sciences

Case study in computational sciences

Cycle of major disciplines

Elective component

Theory of adaptive computational grids

This discipline is devoted to methods of constructing unstructured and structured grids that adapt to certain properties of the area and their use for solving numerical problems in these areas. Such methods of structured grids as methods of equidistribution, Thompson”s method, and such methods of unstructured grids as Delaunay triangulation, Voronoi diagram are considered.

Physics Inferred Neural Networks

Physics Inferred Neural Networks

Generative algorithms

This course introduces students to various generative modeling techniques and algorithms, such as Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and autoregressive models. Students will learn how to design, train, and evaluate generative models, as well as explore their applications in fields like computer vision, natural language processing, and data synthesis.

Course Title: Project Stakeholder Management

The course is designed to study the management of stakeholders (stakeholders) of the project. Undergraduates will consider the basic principles and analysis of the external and internal environment of the project, aimed at identifying and systematizing the main stakeholders, assessing their goals, collecting information about them and using this data in the project management process. It will also consider negotiating and engaging stakeholders to collaborate with managing the expectations of key stakeholders.

IT Project Management

Discipline that focuses on the strategic planning, development, and successful execution of products throughout their lifecycle. It includes a wide range of activities and responsibilities aimed at creating valuable and marketable products that meet customer needs and meet business goals.

Service model in project management

The course covers the area of the service approach in organizing the company’s activities; service tools and services provided by internal divisions and/or external contractors

Heterogeneous Parallelization

This course is a concept, languages, methods and patterns for programming heterogeneous, massive parallel processors. It covers heterogeneous computing architectures, software programming models, memory launching methods and parallel algorithms on the example of CUDA and OpenCl.

Reinforcement learning

This course will introduce students to the basics of reinforcement learning. Upon completion of this course, the student will be able to: Formalize problems as Markov decision-making processes; Understand basic exploration techniques and trade-offs between exploration and exploitation; Understand value functions as a universal tool for making optimal decisions; Know how to implement dynamic programming as an effective approach to solving industrial control problems.

Quantum computations

This discipline involves the study of quantum computing methods and their advantages over classical computation methods. The course covers the main provisions of the classical theory of computational complexity, the gate model of quantum computing, universal sets of gates, quantum computing algorithms based on the quantum Fourier transform, in particular, Shor’s algorithm, quantum search algorithms, algorithms for quantum simulation of physical systems, an introduction to quantum error correction, and error-resistant computations, hybrid quantum-classical algorithms, in particular, variational quantum algorithms.