DCU Home | Our Courses | Loop | Registry | Library | Search DCU

Module Specifications..

Current Academic Year 2023 - 2024

Please note that this information is subject to change.

Module Title Deep Learning
Module Code MS456
School School of Mathematical Sciences
Module Co-ordinatorSemester 1: Vladimir Krylov
Semester 2: Vladimir Krylov
Autumn: Vladimir Krylov
Module TeachersVladimir Krylov
NFQ level 8 Credit Rating 7.5
Pre-requisite None
Co-requisite None
Compatibles None
Incompatibles None
Coursework Only

Deep learning is an emerging branch of machine learning that focuses on learning appropriate data representations from the data itself rather than operating with respect to a predefined model. In the first half of this course we will explore the fundamental techniques and approaches of machine learning that led to the development and success of deep learning. These include stochastic gradient decent, random forests, perceptron, support vector machines and naïve Bayes methods. After that artificial neural networks and the basics of deep learning will be introduced to demonstrate the construction of simple feed-forward and more complex convolutional neural networks. We will address the special type of deep learning architectures that is suitable for time series analysis: recurrent neural networks (RNNs), and in particular, a special type of RNNs - long short-term memory (LSTM) networks. The presentation will focus on actuarial and financial applications to ensure practical understanding of the critical machine learning and deep learning concepts such as representation, generalization, overfitting and model architecture. The lectures will be supported by computer labs demonstrating the use of the proposed methodologies using R / Keras programming language.

Learning Outcomes

1. Solve linearly separable classification problems using trees and their collections
2. Construct kernel-based machine learning methods like Support Vector Machines
3. Apply generative models to performs probabilistic machine learning modeling
4. Apply perceptron and neural networks models
5. Build, train and deploy deep neural networks, RNNs, LSTMs

Workload Full-time hours per semester
Type Hours Description
Laboratory24Computer labs
Independent Study140Independent work
Total Workload: 188

All module information is indicative and subject to change. For further information,students are advised to refer to the University's Marks and Standards and Programme Specific Regulations at: http://www.dcu.ie/registry/examinations/index.shtml

Indicative Content and Learning Activities

Linear models and gradient decent techniques

Decision trees and ensemble learning

Online learning: perceptron, support vector machines

Generative models and naive Bayes methods

Neural networks and backpropagation

Deep learning, RNNs, LSTMs

Assessment Breakdown
Continuous Assessment100% Examination Weight0%
Course Work Breakdown
TypeDescription% of totalAssessment Date
AssignmentAssignment 133%n/a
AssignmentAssignment 233%n/a
Group assignmentAssignment 333%n/a
Reassessment Requirement Type
Resit arrangements are explained by the following categories;
1 = A resit is available for all components of the module
2 = No resit is available for 100% continuous assessment module
3 = No resit is available for the continuous assessment component
This module is category 1
Indicative Reading List

    Other Resources

    Programme or List of Programmes
    ACMBSc in Actuarial Mathematics
    FIMB.Sc. Financial Mathematics

    My DCU | Loop | Disclaimer | Privacy Statement