MACHINE LEARNING

SHE Level 4
SCQF Credit Points 20.00
ECTS Credit Points 10.00
Module Code MHI226720
Module Leader Yan Zhang
School School of Computing, Engineering and Built Environment
Subject Computing
Trimester
  • B (January start)

Summary of Content

This module will provide the fundamental techniques used in Machine Learning and how they can be applied to Data Analysis. It will introduce the basic building blocks of modern Machine learning including Perceptrons and Fully Connected/Convolutional Neural Networks. The Student will be given an introduction into Deep Learning methods to perform such tasks as image classification, image segmentation and natural language processing. After this module students will have sufficient knowledge to be able to format data to be suitable for Machine Learning and be able to successfully train their own AI models with focus on solving relevant real world data science problems.

Syllabus

PRE-REQUISITE KNOWLEDGE: Introduction to Data Science, Probability and Statistics, Decision Trees - K-Nearest Neighbour - K-means - Support Vector Machines - Training - Evaluation - Testing - Performance Evaluation - Perceptrons - Multi-layer Perceptrons - Fully Connected Neural Networks - Convolutional Neural Networks - Back-Propagation - Stochastic Gradient Descent - Activation Functions - Momentum - Learning Rates - Dropout - Overfitting - Hyperparameter Optimisation - Image Classification - Object Detection - Image Segmentation - Text Analysis - Natural Language Processing

Learning Outcomes

On successful completion of the module the student should be able to:1. Understand the principles and goals of Machine Learning 2. Understand the basic techniques used in ML algorithms 3. Be able to select appropriate ML approaches to a given problem 4. Be able to evaluate the effectiveness of an ML algorithm 5. Understand how to train ML algorithms and how to prepare data for training 6. Implement and test ML algorithms in a suitable programming language

Teaching / Learning Strategy

Lectures are supplemented by directed reading to relevant sources both hard and electronic format and varied further reading is encouraged. Hands on experience is gained in the process of planning simulated projects. Students are supported in their studies by both face-to-face and on-line tutorials and online quiz material. Student assessment will be based around on-line techniques for three of the assessments and a class based test for the remaining assessment. Learning and teaching strategies will be developed and implemented, appropriate to students' needs, to enable all students to participate fully in the module.

Indicative Reading

Pattern Recognition and Machine Learning by Christopher M. Bishop Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig Deep Learning (Adaptive Computation and Machine Learning Series) by Ian Goodfellow , Yoshua Bengio

Transferrable Skills

Critical thinking and problem solving (D1) Time management (organising and planning work) (D6) Information retrieval skills (D10) IT Skills (D13) Communication skills, written, oral and listening (D14)

Module Structure

Activity Total Hours
Lectures (FT) 24.00
Independent Learning (FT) 100.00
Assessment (FT) 28.00
Tutorials (FT) 24.00
Seminars (FT) 12.00

Assessment Methods

Component Duration Weighting Threshold Description
Course Work 01 n/a 50.00 35% Class based test
Course Work 02 n/a 50.00 35% Class based test