Module: Machine Learning for Physics

Latest Module Specifications

Current Academic Year 2025 - 2026

Module Title	Machine Learning for Physics
Module Code	PHY1091
Faculty	Science & Health	School	Physical Sciences
NFQ level	8	Credit Rating	5

Description

This module will provide students both with a strong theoretical grounding and with practical hands-on experience of modern machine-learning methods and their application to problems in the physical sciences. Students will learn the underlying principles behind key algorithms, from simple classification tools to deep neural networks, and how to implement these practically using common platforms such as Scikit-Learn and TensorFlow. Real world datasets from Physics research papers will be used to illustrate machine-learning methods throughout the module. The module will also explore areas of overlap between physics and the development of new machine-learning technologies, focussing on the use of ML to advance physics research and the use of physical principles to improve both ML algorithms and the hardware required to run them.

Learning Outcomes

1. Explain the key concepts and applications of common supervised and unsupervised Machine Learning methods.
2. Explore a physics data set and implement the appropriate data-preprocessing for Machine Learning methods.
3. Select and justify an appropriate machine learning technique for a given problem.
4. Demonstrate knowledge of, and ability and use, appropriate approaches and existing ML tools (such as Scikit-learn and Tensorflow/Keras) to develop, manage, and evaluate ML projects and solutions in the field of Physics.
5. Explain the concepts of different types of Artificial Neural Networks and their application to both general and physics-related problems
6. Identify, explain and extrapolate from emerging trends in the domains of Machine Learning and Physics, including Physics-informed ML and Neuromorphic Computing.

*Type*	*Hours*	*Description*
Workload	Full time hours per semester
Lecture	18	Core lecture material
Laboratory	8	Computer-based assignments
Assignment Completion	14	Students will work on various assignments throughout the module
Independent Study	85	Students will study the module material, related resources and references
Total Workload: 125

Section Breakdown
CRN	21168	Part of Term	Semester 2
Coursework	50%	Examination Weight	50%
Grade Scale	40PASS	Pass Both Elements	N
Resit Category	RC1	Best Mark	N
Module Co-ordinator	Stephen Power	Module Teacher

Type	Description	% of total	Assessment Date
Assessment Breakdown
Assignment	Coding problems involving the practical application of machine-learning concepts to physics data sets	40%	n/a
Assignment	Take-home problems to be submitted using Loop	10%	n/a
Formal Examination	End-of-Semester Final Examination	50%	End-of-Semester

Reassessment Requirement Type
Resit arrangements are explained by the following categories; RC1: A resit is available for both^* components of the module. RC2: No resit is available for a 100% coursework module. RC3: No resit is available for the coursework component where there is a coursework and summative examination element. ^* ‘Both’ is used in the context of the module having a coursework/summative examination split; where the module is 100% coursework, there will also be a resit of the assessment

Pre-requisite	None
Co-requisite	None
Compatibles	None
Incompatibles	None

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

Key Concepts, Tools and Methods
Primer / refresher on coding and mathematical tools (numpy, matplotlib, pandas, linear algebra, partial differentiation).

Outline of a machine-learning pipeline.

Data exploration and preprocessing.

Supervised and Unsupervised Learning
Common classification and regression algorithms (decision trees, kNN, support vector machines).

Applying and evaluating ML algorithms in python using scikit-learn.

How to prepare physics data sets and apply appropriate ML methods to make predictions.

Clustering methods and applications in physics.

Artificial Neural Networks and Deep Learning
Comparison of real and artificial neural networks.

Components and mathematics of an artificial neural network (neurons, weights, biases, activation functions).

Hyperparameters and training.

Convolutional neural networks for image analysis.

Implementing NNs in python using keras and tensorflow.

Applications of NNs to problems in physics.

Advanced Topics
Neural networks for time series and language.

Autoencoders and Generative AI.

Reinforcement Learning.

Physics-informed ML.

Architectures and technologies for Machine Learning.

Future directions (Neuromorphic computing, Quantum ML).

Indicative Reading List

Books:

Viviana Acquaviva: 2023, Machine Learning for Physics and Astronomy, Princeton University Press, 280, 9780691206417
Aurélien Géron: 0, Hands-On Machine Learning with Scikit-Learn, Keras, and Tensorflow, 3rd Ed., 9781098125974
Francois Chollet: 2021, Deep Learning with Python, Second Edition, Simon and Schuster, 502, 9781617296864

Articles:
None

Other Resources

None

New 4th year module delivered by School of Physical Sciences in 2nd semester of year 24/25. Will be core for PAN4 and optional for AP4, PBM4 and PHA4.