Probabilistic modelling, learning and algorithms
Duration
2-5 h
Syllabus
- Motivation: supervised and unsupervised learning, topic models
- The EM algorithm
- Gibbs sampling
- Variational inference
- Hyperparameter learning
- Practical topic modelling
- Some software implementations
Prerequisites
Some statistical and Python background
Credential
Taught in executive (ad-hoc) courses, and also part of ‘Statistical modelling and inference’ in MSc Data Science