Full Title Machine Learning

Short Title Machine Learning

Code COMP09012
Level 09
Credit 05

Author Mullery, Sean
Department Mech. and Electronic Eng.

Subject Area Computing
Attendence N/A%
Fee

Description

This module introduces the topic of machine learning algorithms (algorithms that learn from data), with the first part of the module dedicated to the standard shallow forms of machine learning before moving on to Deep Learning and Convolutional Neural Networks for use in computer vision tasks, particularly recognition, classification and localisation. The emerging topic of Deep Reinforcement Learning will be briefly introduced. The module will look at training strategies and frameworks for Deep Learning. As well as the technical/scientific elements, students will reflect on the ethical implications of machine learning.


Indicative Syllabus

Machine Learning: Logistic Regression, KNN, SVM, kernel SVM, Decision Trees.

Principle Component Analysis (PCA)

Model evaluation and Hyperparameter Tuning.

Deep Learning: Perceptron, multi-layer perceptron, representation learning.

Back propagation, SGD (variants), Cost/loss functions, regularisation.

Convolutional Neural Networks.

Dropout, Batch-normalisation.

Image classification, image recognition, localisation.

Transfer Learning.

Introduction to Reinforcement Learning.

Data Sets, ImageNet, Kaggle.

Data distributions, bias, variance, divergence (KL, JS, Wasserstein Distance).

Machine Learning Frameworks such as TensorFlow, Keras, PyTorch and Caffe.

Citizen rights under data protection, data storage.


Learning Outcomes
On completion of this module the learner will/should be able to
  1. Compare state of the hand engineered detectors with machine learning techniques in terms of performance on appropriate metrics and data sets and determine the appropriateness of each for safety critical applications.

  2. Apply transfer learning to adapt a pre-trained network to a new classification problem.

  3. Assess the validity of various cost functions to specific machine learning problems.

  4. Effectively collaborate and communicate with others in the timely development of solutions to machine learning problems, including reports and software.

  5. Design, test and evaluate deep network architectures.

  6. Appreciate the data rights of citizens and the constraints these apply to the use of pattern detection in real world scenarios.


Assessment Strategies

A terminal exam and continuous assessment in the form of group project work will be used to assess the module.

To reinforce the theoretical principles covered in lectures, learners will participate in project work.

The learner will complete a final exam at the end of the semester.

The learner is required to pass both the projects and terminal examination element of this module.

 


Module Dependencies
Pre Requisite Modules
Co Requisite Modules
Incompatible Modules

Coursework Assessment Breakdown %
Course Work / Continuous Assessment 60 %
End of Semester / Year Formal Examination 40 %

Coursework Assessment Breakdown

Description Outcome Assessed % of Total Assessment Week
Individual Project 3,5 30 Week 6
Group Project 2,3,4,5 30 Week 12


End Exam Assessment Breakdown

Description Outcome Assessed % of Total Assessment Week
Terminal Exam 1,3,5,6 40 End of Semester


Mode Workload

Type Location Description Hours Frequency Avg Weekly Workload
Lecture Lecture Theatre Lecture 2 Weekly 2.00
Laboratory Practical Computer Laboratory Laboratory Practical 2 Fortnightly 1.00
Independent Learning Not Specified Independent Learning 7 Weekly 7.00

Total Average Weekly Learner Workload 3.00 Hours

Mode Workload

Type Location Description Hours Frequency Avg Weekly Workload

Total Average Weekly Learner Workload 0.00 Hours

Mode Workload

Type Location Description Hours Frequency Avg Weekly Workload

Total Average Weekly Learner Workload 0.00 Hours

Mode Workload

Type Location Description Hours Frequency Avg Weekly Workload
Lecture Online Lecture 2 Weekly 2.00
Independent Learning Not Specified Independent Learning 7.5 Weekly 7.50
Laboratory Practical Online Online Lab 0.5 Weekly 0.50

Total Average Weekly Learner Workload 2.50 Hours

Resources
Book Resources

Other Resources

x

Url Resources

x

Additional Info

ISBN BookList

Book Cover Book Details
Ian Goodfellow 2017 Deep Learning (Adaptive Computation and Machine Learning Series) MIT Press
ISBN-10 0262035618 ISBN-13 9780262035613
Francois Chollet 2018 Deep Learning with Python Manning Publications
ISBN-10 1617294438 ISBN-13 9781617294433
Christopher M. Bishop 2007 Pattern Recognition and Machine Learning (Information Science and Statistics) Springer
ISBN-10 0387310738 ISBN-13 9780387310732