Introductory description

The module introduces key concepts related to Machine Learning (ML) and Artificial Intelligence (AI). Through exploring Linear and Logistic Regression techniques, students will embark on learning critical skills to tackle more advanced Neural Networks architectures. Key focus will be made on data management workflow to enhance performance and robustness of ML algorithms. Relevant AI techniques that are widely used in development of automated vehicles and related areas will be explored.

Module aims

Outline syllabus

This is an indicative module outline only to give an indication of the sort of topics that may be covered. Actual sessions held may differ.

Indicative content

• Introduction to Regression and classification problems:
• Linear regression
• Logistic regression
• Deep learning systems
• Introduction to neural networks
• Convolutional neural networks
• Recurrent neural networks
• Reinforcement learning

Learning outcomes

By the end of the module, students should be able to:

• Develop and optimise neural network architectures, including convolutional and recurrent models. [AHEP:4, M1, M2, M3, M4, M5, M7]
• Demonstrate advanced proficiency in the application of machine learning algorithms by designing data pipelines, selecting suitable optimisation techniques, and conducting model performance analysis. [AHEP:4, M1, M2, M3, M4]
• Propose solutions to improve model generalisation and address common challenges. [AHEP:4, M1, M2, M3, M4, M5, M7]
• Critically evaluate advanced machine learning concepts to design, implement, and assess predictive models for complex data-driven tasks. [AHEP:4, M1]

Indicative reading list

Reading lists can be found in Talis

Specific reading list for the module

Subject specific skills

1. Machine Learning Implementation: Apply supervised, unsupervised, and reinforcement learning algorithms to solve data-driven problems.

2. Neural Network Design: Develop and train architectures such as Convolutional Neural Networks, Recurrent Neural Networks, and Gated-Recurrent Neural Networks for tasks like classification and time-series forecasting.

3. Data Preprocessing and Augmentation: Prepare datasets by cleaning, transforming, and augmenting data to enhance model performance.

4. Model Evaluation: Use metrics like accuracy, precision, recall, F1-score, and confusion matrices to assess model effectiveness.

5. Optimisation Techniques: Implement gradient-based optimisation algorithms, such as stochastic gradient descent (SGD), to train models effectively.

6. Programming with Python and PyTorch: Write code in Python and use PyTorch for building and training machine learning models.

Transferable skills

1. Critical Thinking and Problem-Solving: Evaluate complex situations and develop solutions by applying analytical reasoning.

2. Project Management: Manage end-to-end machine learning workflows, from problem formulation to solution deployment.

3. Data-Driven Decision Making: Interpret and derive actionable insights from data analysis.

4. Collaboration and Teamwork: Work effectively with others during coding tasks or problem-solving exercises.

5. Communication Skills: Convey technical concepts, results, and insights clearly through reports, presentations, and documentation.

6. Time Management: Plan and prioritise tasks to meet deadlines in coding projects and assignments.

7. Technical Literacy: Gain proficiency with modern tools and platforms like Python libraries (e.g., TensorFlow, PyTorch) and Jupyter notebooks.

8. Adaptability: Apply learning to unfamiliar datasets and rapidly evolving technology trends in machine learning and AI.