»Ê¼Ò»ªÈË

This BASc module covers the basic principles of machine reasoning, exploring the foundations of the rapidly developing field of artificial intelligence, and outlining the mathematical techniques used in both knowledge representation and future artificial intelligence modules.

The first part of the module will introduce the mathematical and logical theories used in the development of machine reasoning and knowledge representation. The students will learn the notions of agent-based systems (e.g., intelligent agents, problem-solving agents, knowledge-based agents), the syntax, semantics, and use of first-order logic in knowledge representation and inference systems. A modal logic for AI will be briefly presented as an overview on more advanced topics in knowledge representation.

The second part of the module will focus on learning and reasoning under uncertainty by using probabilistic techniques (e.g., Naïve Bayes models and Bayesian networks, introductory hidden Markov models) as an introduction to expert systems and machine learning.

Once equipped with the main technical and theoretical tools, the students will be presented with a selection of different applications of machine reasoning, e.g., natural language processing, machine vision, and robotics, to create a point of contact with real-world examples and future, more advanced AI modules.

Teaching Delivery

This module is taught in 10 weeks lectures and 10 PGTA led seminars.

Indicative Topics

• Introduction to mathematical and logical theories used in the development of machine reasoning and knowledge representationÌý
• Ìýagent-based systems (e.g., intelligent agents, problem-solving agents, knowledge-based agents),ÌýÌý
• the syntax, semantics, and use of first-order logic in knowledge representation and inference systems.Ìý
• modal logic for AI (theory only)Ìý
• learning and reasoning under uncertainty by using probabilistic techniques (e.g., Naïve Bayes models and Bayesian networks, introductory hidden Markov models)ÌýÌý
• expert systems and machine learning.Ìý
• applications of machine reasoning, e.g., natural language processing, machine vision, and robotics

Module aims and objectives

• Be able to discuss first order logic in the context of knowledge representationÌý
• Discuss and solve exercises with agent based systemsÌý
• Describe the concept of inference in machine learningÌý
• Use probabilistic techniques to solve exercises in machine learning domainÌý
• Discuss the main components of Machine Learning, Natural Language Processing, Machine VisionÌý