Advanced Machine Learning and Deep Learning Techniques for Business and Economics
The module examines recent developments in supervised and unsupervised machine learning, including XGBoost, CatBoost, LightGBM and SHAP value analysis. Students are also introduced to deep neural networks, recurrent and convolutional networks, reinforcement learning, natural language processing and generative models, with an emphasis on transparent inference and practical problem-solving.
Prerequisites
To take this module you must first have completed Foundations of Econometrics, Foundations of Data Science and Programming in Python.
Delivery
Live online lectures and tutorials, conducted at scheduled times and recorded for review.
Study Hours
Approximately 200 hours of study, of which 25 are contact hours and 175 are private study.
Software Used
Python, XGBoost, CatBoost, LightGBM
Credits
20 CATS credits
Programme Faculty
Key Outcomes
- —Apply advanced machine learning methods to complex problems
- —Develop deep learning and reinforcement learning models
- —Use ensemble methods and SHAP values to interpret predictions
- —Understand natural language processing and generative models
- —Communicate complex analytical arguments clearly
Lecture Programme
- WK 01
Introduction to advanced machine learning
- ·Advanced supervised and unsupervised methods
- ·Transparent inference and SHAP values
- WK 02
Ensemble methods
- ·XGBoost
- ·CatBoost
- ·LightGBM
- ·Recommendation systems
- WK 03
Deep learning
- ·Deep neural networks
- ·Recurrent neural networks
- ·Convolutional neural networks
- WK 04
Reinforcement learning and advanced applications
- ·Deep reinforcement learning
- ·Natural language processing
- ·Generative models
Assessment
Online coursework submission with support from the module lecturer and tutor.