ELEC60019 Machine LearningLecturer(s): Dr Abdalrahman Abu Ebayyeh; Dr Deniz Gunduz Aims
Upon successful completion of this module, you will be able to:
1. Develop solutions to machine learning problems by modelling and preprocessing data, and designing, selecting and develop appropriate learning algorithms. 2. Consider and contrast the problems of learning and overfitting in an ML system 3. Jutsify the use of linear regression, classification, logistic regression, support vector machines, neural networks, nearest neighbour and clustering. 4. Recommend and construct the use of a machine learning algorithm in unseen situations. Learning Outcomes
Upon successful completion of this module, you will be able to: 1. Solve a machine learning problem by modelling and preprocessing data, and designing, selecting and implementing appropriate learning algorithms. 2. Explain and contrast the problems of learning and overfitting. 3. Apply machine learning algorithms in unseen situations. 4. Use and explain linear regression, classification, logistic regression, support vector machines, neural networks and reinforcement learning.
Syllabus
Part 1. Components of learning, tasks, types of learning, ML problem formulation,simple predictors
Part 2. Feasibility of learning, error function, Empirical Risk Minimization, generalisationbounds, performance vs complexity, bias/variance trade off, Hoeffding/VC inequalities Part 3. Feature transformations, noisy data, overfitting, regularisation Part 4. Logistic regression, gradient descent, Perceptron, Multi Layer Perceptron,Neural Network, backpropagation Part 5. Hyperplane, separation with hard margin, soft margin, support vector machines, Part 6. Nearest neighbour classification, linear unsupervised learning, principlecomponent analysis Part 7. Kmeans clustering, kernel Kmeans, advanced clustering algorithms Exam Duration: 3:00hrs Exam contribution: 80% Coursework contribution: 20% Term: Autumn Closed or Open Book (end of year exam): N/A Coursework Requirement: N/A Oral Exam Required (as final assessment): N/A Prerequisite module(s): None required Course Homepage: Blackboard Book List:
