Courses_
Modern Statistical Learning
Machine learning from a statistical perspective
This two-day workshop aims to provide an introduction to machine learning language and methods to applied statisticians and health researchers involved in statistical analysis.
Course overview
Recent years have brought a rapid growth in the amount and complexity of data in biostatistical applications. Among others, data collected in imaging, genomic, health registries, call for new statistical techniques in both predictive and descriptive learning. Machine learning algorithms for classification and prediction, complement classical statistical tools in the analysis of these data.
This workshop will provide a gentle introduction and overview of many of these new methods such as, classification trees, random forests, model selection, lasso, bootstrapping, cross-validation, gam, splines, and many others. The presentation will focus on concepts rather than on the mathematical details and there will be opportunity to practice the use of these methods in R.
You'll work through seven topics including:
- Basics of R
- What is Statistical Learning?
- Regression and Classification
- Resampling methods
- Regularization and model selection
- Beyond linearity
- Beyond additivity
Please note:
Lunch and morning tea will be offered.
| Key information | |
|---|---|
| Course fees |
|
| Course structure |
|
| Course dates |
|
| Application deadlines |
|
| Assumed knowledge |
Basic understanding of classical statistical techniques such as linear and logistic regression. Familiarity with R is helpful but not essential. There will be an introductions to R session (1h) before the workshop. |
| Computer and software | We will use R. Computers are available in the room but you can bring your own. |
How to apply
This course is now full. For any enquires contact sph.bsta@sydney.edu.au.
