Artificial Intelligence

Entry requirements: C#/C++; Algorithms and Data Structures; Design Patterns

Credits: 3

Course: Core

Language of the course: Russian

Lecturer

Sergey Kovalchuk

Objectives

  • To understand the tasks of Game Artificial Intelligence;
  • To learn algorithms of configuration and findpath in ill-defined spaces, of analyzing and avoiding threats, as well as crowd simulation technologies;
  • To configure Game Artificial Intelligence
  • To master the skills of testing, profiling and optimizing AI algorithms

Contents

The main goal of artificial intelligence (except for special cases) is to not to destroy the player but entertain the one. The delicate balance of suddenness and predictability and even more delicate exquisite boundary between good game and loss are the aspects maintaining the interest of the player to the game covered by the course. We will teach you how to create required behaviors of the characters controlled by artificial intelligence, threat analysis and the environment, the relationship with the character animation as well as interaction of game narrative and artificial intelligence.

Basics of Artificial Intelligence: Intelligent technologies. Intuitive approach. Symbol modelling. Knowledge paradigm. Modelling of thought. Discrete choice. Robotics.
Reasoning with uncertainty: Uncertainty and incompleteness. Probabilistic reasoning. Fuzzy logic. Bayesian networks. Neural networks. Confidence.
Methods of data mining: Data mining and machine learning methods. Parametric and non-parametric knowledge models. Rules with computable coefficients. Extracting knowledge from natural language texts.
Methods of knowledge formalization and interpretation: Semantic networks. Ontology-based relations. Production systems. Knowledge representation models. Frames.
Reasoning and decision making: Forward and backward reasoning. Rule-based reasoning. Case-based reasoning. Hybrid reasoning. Reasoning quality evaluation. Increasing reasoning.
Adaptation and self-learning: System memory. Dynamic data mining methods. Configuration and evaluation of knowledge models. Learning from observation.
Expert and intelligent systems: Decision support system technologies. DSS architecture. Expert systems. Rule-based systems. Data-based systems. Model-based systems. DSS development trends.
IA in finance: DSS for finance. Finance recommendation systems. Target acquisition. Stock market robots. User support automation. Intelligent customized interfaces.

Format

Lectures and labs. Practical sessions.

Assessment

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