The MLx Fundamentals Course aims to offer a strong foundation in Machine Learning through a comprehensive training program that includes theory and practical sessions covering key areas, including:
​
-
Mathematics of ML, and fundamentals of statistical/probabilistic ML (e.g., linear algebra, probability theory, linear models, SVM, tree-based models, and other advanced models for tabular data).
​
-
Factorisation techniques (from PCA, ICA, NMF, and tensor factorisation techniques, to their latest advanced counterparts (e.g., neural, and probabilistic versions), these techniques have shown to be an effective tool in recommender systems, signal processing and more).
​
-
Advanced practical topics in ML (e.g., generalisation, robustness, dealing with missing data, and several other essential topics for real-world ML and data science applications/problems).
​
-
Optimisation (e.g., convexity, gradient methods, non-convex optimisation, and other topics that are essential in understanding, using, and implementing the latest ML and DL techniques).
​
-
Fundamentals of representation learning (e.g., basic neural net architectures, activation and loss functions, backprop, and other essential topics in representation learning)
​
-
Generative AI – Vision (reviewing the DL techniques in computer vision, including, the latest generative AI techniques for images and videos).
​
-
Generative AI – Language (reviewing the DL techniques in NLP, including, LLMs and the latest generative AI techniques for text and sequences).
​
​
This course will have four 1.5hr TA-supported practical sessions, including coding practicals, in the following areas:
-
Statistical ML
-
DNN + Optimisation
-
Gen. AI in vision
-
Gen. AI in language
​