Q3 or Q4, 2020

Danone Nutricia Research, Utrecht

Presented by

Jan Willem Bikker and/or Bert Schriever (CQM)

Price*: €200

* Including lunch, coffee & tea.

Registration postponed until further notice due to developments in the outbreak of nCov-19 (novel Coronavirus)

At the moment, the expectations from Machine Learning and Artificial Intelligence techniques are high. Data scientists applies these techniques frequently, but also statisticians come across them more often. Therefore, as a statistician, is it very useful to have an understanding of these modern Data Science techniques. If you are curious and want to know more, then please join our course!

In this short course, some well-known ML techniques and their applications to solve questions will be introduced and explained from a statistical perspective by presentations and discussion.

As such, participants are expected to have a good basic understanding of classical statistical techniques.

Due to time constraints, techniques can’t be explained in full detail and practical exercises are not part of this 1-day course.

Preliminary Program

09:30

Registration, coffee & tea

10:30

Introduction – Why Machine Learning
Statisticians’ view on some aspect of ML

11:15

Coffee Break

11:30

Basics of most popular Machine Learning Techniques and Terminology

  • Regression techniques, like Ridge and Lasso regression
  • Classification techniques, like Linear and Polynomial Support Vector Machines, Classification Trees (CART, Random Forest)
  • “No free-lunch” theorem

12:45

Lunch break

13:30

Continuation of ML techniques

16:15

Coffee break

16:30

Generalisation

  • Train, Validate, Test set, Cross Validation and Evaluation of a model
  • Typical issues in Machine Learning projects
  • Beyond prediction: causal inference, Sampling bias, n < p (more predictors than observations), Missing observations and Outliers

18:00

Closure