To review different learning paradigms and artificial intelligence techniques including deep learning.

To understand basic concept and theory of artificial intelligence, neural networks and machine learning.

To acquire knowledge of intelligent optimisation techniques such as genetic algorithm, swarm-based algorithms including the Bees Algorithm.

To acquire knowledge of fuzzy logic and its use in control and other applications.

To apply the above techniques to solve real world problems.

1. Appreciate the concept of artificial intelligence techniques including neural networks and their applications in industry.

2. Comprehend what machine learning is and how it could be used in the industry.

3. Assess the methods of intelligent and swarm-based optimisation methods and their practical use.

4. Appreciate fuzzy logic and its implementation to deal with control and decision making.

5. Systematically apply the techniques of deep learning to real world problems and reflect on their effectiveness.

The module will be delivered through a blend of online teaching and learning material, guided study, and on-campus face-to-face classes (tutorials, practical sessions).

This module will enable you to develop your academic skills to think independently and critically and in particular for:

• Employing artificial intelligence techniques for designing solutions for real applications including manufacturing.
• Data modelling, prediction, pattern recognition, classification and clustering using neural networks.
• Data analysis and knowledge creation using machine learning techniques.
• Finding optimum solutions using genetic algorithm and swarm based algorithms.
• Implementing control solutions using fuzzy logic.
• Applying deep learning methods.
• Relating the knowledge and skills obtained from the course to real-life industrial situations in order to obtain viable engineering solutions.

The module will be assessed through two summative components namely by 50% Exam and 50% Coursework.

There is a potential for re-assessment in this module which may result in resitting the failed components.

Coursework allows students to demonstrate learning outcomes 1, 2 and 5. The written exam will cover all the learning outcomes.

• Artificial Intelligence
  • Gradient Descent (GD) algorithm
  • Back Propagation (BP) algorithm
  • Least squares (LS) algorithm
• Machine Learning
  • Supervised/ Unsupervised /reinforcement learning
• Neural Networks
  • Multi-layer perceptron(MLP) networks
  • Kohonen Self Organising Map (KSOM) networks
  • Learning Vector Quantisation (LVQ) networks
  • Adaptive Resonance Theory (ART) networks
  • Radial basis function (RBF) networks
  • Group method of data handling (GMDH) networks
• Genetic Algorithms and Bees Algorithm
• Fuzzy Logic control
  • Logic operations, Fuzzification, Defuzzification, Fuzzy control
• Deep learning
  • Challenges motivating deep learning, Convolutional Networks, Efficient convolution algorithms, Sequence Modeling: Recurrent and Recursive Nets, Practical Methodology and Applications