Detail
Veranstaltungsdetails | Machine learning
Code | Startdatum | Enddatum | Ort | Organisation/Kontakt | OT-SC-WS-03| 20.10.-21.10.2021 (9:00-16:00; on each day) | Online via VC | 20.10.2021 | 21.10.2021 | Online | Data Train data-trainprotect me ?!vw.uni-bremenprotect me ?!.de |
---|
Freie Plätze | Max. Teilnehmer | 1 | 24 |
---|
Beschreibung
Nowadays, machine learning is everywhere: old and new questions, problems and challenges are tackled with machine learning, sometimes with great success, sometimes not. To successfully use machine learning and understand its limitations, we have to go beyond buzzword bingo and learn the general and basic concepts of machine learning.
Learning contents
This workshop teaches the major concepts of machine learning. We focus on general principles instead of going into full depth of single methods. We will learn the difference between supervised and unsupervised learning, important notation such as models and learners, why training errors are different than test errors and how to optimize and evaluate prediction performance. Nevertheless, the course covers the most important machine learning methods such as k-nearest neighbors, decision trees, random forests, boosting, support vector machines and artificial neural networks. The methods will be introduced in a non-technical and intuitive way. These theory sessions will be complemented by hands-on sessions in R, where the methods are applied in practice.
Learning outcomes
Understand basic concepts of machine learning:
- Supervised and unsupervised learning
- Difference between models and learners, training and test errors, etc.
- Over- and underfitting
- Hyperparameter tuning
- Performance evaluation
Know the major machine learning methods:
- K-Nearest Neighbors
- Decision Trees
- Random Forests
- Boosting
- Support Vector Machines
- Artificial Neural Networks
Be able to perform machine learning analyses in R:
- Model fitting
- Hyperparameter tuning
- Performance evaluation
- Benchmarking
Visualizing results
Prior knowledge
Advances math is generally NOT required (only for short part on support vector machines); Basic programming skills are required.
Technical requirements
Computer with access to online course platform (RStudio) or with R installed
(for online format a second screen is advantageous)
Further reading
General introduction:
- James, Witten, Hastie, Tibshirani (2010). An Introduction to Statistical Learning. MIT Press, http://www-bcf.usc.edu/~gareth/ISL/
Advanced:
- Hastie, Tibshirani, Friedman (2009). The Elements of Statistical Learning. Springer, https://web.stanford.edu/~hastie/ElemStatLearn/.
- Both books are freely available on the websites.
Additional comments
---
Programm/Ablauf
Wednesday, 20 October 2021 | 9:00-16:00 |
Thursday, 21 October 2021 | 9:00-16:00 |