Computing professionals are required to solve increasingly complex problems using new algorithms, systems or programming paradigms. Fortunately, black box computational intelligence tools can be configured and applied to problems without revealing intimate knowledge of low-level details to a user.

This course is for computational professionals who are interested in exploring new techniques for solving problems that are ill-defined, have conflicting constraints, or contain data with high noise levels.

Students will discover the industrial applications found in software algorithm development, electronic design automation, data mining, medical diagnosis, and pattern matching.

You will learn the strengths and weaknesses of various computational and artificial intelligence (AI) techniques using supplied software. There is also a brief introduction to spiking neural networks, which uses more sophisticated, and more capable, neuronal models and networks to address problems usually attempted by traditional neural networks.

Learning Outcomes
At the conclusion of the course, you should be able to

• Determine if a particular task is suitable for a computational intelligence technique
• Evaluate the performance of different computational intelligence techniques in solving real-world problems and choose the most appropriate technique for a given problem
• Develop solutions for optimization problems using common algorithms and techniques used in computational and artificial intelligence

Topics Include:

• Search spaces and their importance for assessing problem complexity
• Evolutionary computation, the fundamental engine behind many AI techniques
• Genetic programming (with many examples)
• Neural networks and the iris problem
• Swarm intelligence, the power of collective, decentralized systems
• Support vector machines: a demonstration using a popular tool for simple classification
• Fuzzy logic, including a solution of the traveling salesman problem
• Spiking neural network introduction

You will learn to solve AI problems using software provided as an ISO file which can be loaded into VirtualBox, enabling you to learn techniques for representing and structuring real-world problems using AI. By the end of the course, you will understand common algorithms and techniques used to solve real-world optimization problems, and also gain experience applying them to practical problems.

Skills Needed: Experience with a computer programming language and basic algebra skills.