The course will cover selected topics of advanced statistics and machine learning. Lectures on some topics will be accompanied with presentations by participants, “Excursion” talks on applications in research, and basic practicals in the afternoon. The course requires basic knowledge of statistics. The practical session require basic knowledge with a programming language – examples will be provided in R.

1.3 Presentations by participants (mandatory for assignment)

Participants will give a presentation (20min + 10min Q&A) on a paper or topic of their choice. Below you can find a list of suggested papers. If you want to work on a topic in a team of 2 (i.e. 40min+20min Q&A) or suggest an alternative topic please inquire this until 31st October with the proposed topic to mjung@bgc-jena.mpg.de.

During registration please choose a topic that was not yet chosen.

All presentations need to be ready on Monday 16th Nov 2020 at 9 am. The detailed schedule will be announced then.

The presentations should be educational and try to focus on the important things one should know about a method when applying it, i.e. the principle, advantages, disadvantages, assumptions, and pitfalls, rather than all mathematic details, derivations, theorems and proofs. Practical examples are often very illustrative.

1.4 Other Preparations

Bring a laptop with a recent version of R being installed or running for the practicals. If you prefer another language, that is fine but we will not provide corresponding code examples. Please also make sure that you can access the internet via WLAN (BGC-users, if you have a BGC-account; BGC-guests, if you don't have an account).

1.5 Preliminary agenda

Day	Topic	Who
Monday, November 16
9:00 - 09:45	Introduction to basic statistical tools	Martin Jung
09:45 - 10:00	Break
10:00 - 10:30	Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy
10:30 - 11:00	Toward the true near‐surface wind speed: Error modeling and calibration using triple collocation
11:00 - 11:15	Break
11:15 - 12:00	Hierarchical Clustering via Joint Between-Within Distances: Extending Ward's Minimum Variance Method
12:00 - 13:00	Lunch Break
13:00 - 13:30	Archetypal Analysis
13:30 - 14:00	Visualizing Data using t-SNE
14:00 - 14:15	Break
14:15 - 15:00	Dimensionality reduction	Mirco Migliavacca
15:00 - 17:00	Practical	Mirco Migliavacca
Tuesday, November 17
9:00 - 09:45	Time series analysis	Lina Estupinan-Suarez
09:45 - 10:00	Break
10:00 - 10:30	Cross-validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure
10:30 - 11:00	Summarizing multiple aspects of model performance in a single diagram
11:00 - 11:15	Break
11:15 - 12:00	EXCURSION	Nora Linscheid
12:00 - 13:00	Lunch Break
13:00 - 13:30	BGI SEMINAR
13:30 - 14:00	BGI SEMINAR
14:00 - 14:15	Break
14:15 - 15:00	Mixed effect model	Thomas Wutzler
15:00 - 17:00	Practical	Thomas Wutzler
Wednesday, November 18
9:00 - 09:45	Random Forests	Martin Jung
09:45 - 10:00	Break
10:00 - 10:30	Bias in random forest variable importance measures: Illustrations, sources and a solution
10:30 - 11:00	Isolation Forest
11:00 - 11:15	Break
11:15 - 12:00	EXCURSION	Jacob Nelson
12:00 - 13:00	Lunch Break
13:00 - 13:30	A working guide to boosted regression trees
13:30 - 14:00	A unified approach to interpreting model predictions
14:00 - 14:15	Break
14:15 - 15:00	Model evaluation	Martin Jung
15:00 - 17:00	Practical	Simon Bessnard
Thursday, November 19
9:00 - 09:45	Neural Networks	Basil Kraft
09:45 - 10:00	Break
10:00 - 10:30	Deep learning
10:30 - 11:00	Long Short-Term Memory
11:00 - 11:15	Break
11:15 - 12:00	EXCURSION	Basil Kraft
12:00 - 13:00	Lunch Break
13:00 - 13:30	Variable Importance	Martin Jung
13:30 - 15:00	Practical	Fabian Ganz
Friday, November 20
9:00 - 09:45	Parameter estimation	Nuno Carvalhais
09:45 - 10:00	Break
10:00 - 10:30	Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling
10:30 - 11:00	A comparison of techniques for the estimation of model prediction uncertainty
11:00 - 11:15	Break
11:15 - 12:00	EXCURSION	Tina Trautmann
12:00 - 13:00	Lunch Break
13:00 - 13:30	Deep learning and process understanding for data-driven Earth system science
13:30 - 14:00	Feedback

1.6 Interested?

Prerequisites:

Basic knowledge of a language of scientific computing: R, Matlab
Make use of the R course - The basics
Either the course 'Basic statistics' or recalling the typical “statistics 1” type of lectures from university.

Exercises will be in R – the use of any other language is welcome; however support depends on the person in charge and cannot be guaranteed.

Learn R… Here is a list of useful online resources to help you bring your R skills to a new level.
The material from the R basics course might also be useful for you.

1.7 Material

Here, you can download the papers, which you will need for your presentation.

1.8 Requirements for the assignment

All participants have to prepare a short presentation on one "unconventional" method of their choice: Every day will have a few of these presentations and we want to discuss with you about the pros and cons: Please register for one of the following topics (but feel free to add another one).

Important

Don’t choose a technique that you know already!
Check the list of participants below and choose a topic that has not yet been selected. Ideally, we would like to cover all topics.

....and note that we are not necessarily experts in the methods.

# / NAME OF PRESENTER	Topic	Context
1 SOPHIA WALTER	Archetypal Analysis	Multivariate data representation
2 / ANN-SOPHIE LEHNERT	A working guide to boosted regression trees	non parametric regression
3 /	From outliers to prototypes: Ordering data	novelty/outlier detection
4 / SANTIAGO BOTIA	Long Short-Term Memory	neural networks for time series
5	Calibration of process-oriented models	model calibration and evaluation
6 / SOPHIE VON FROMM	Deep learning	deep learning overview
7 / CAGLAR KUCUK	A unified approach to interpreting model predictions	variable importance, explainable AI
8	Quantile regression forestsa	random forest, quantile regression
9 /	Deep learning and process understanding for data-driven Earth system science	deep learning and hybrid modeling for Earth System Science
10 / SINIKKA PAULUS	Cross-validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure	model evaluation
11	MissForest—non-parametric missing value imputation for mixed-type data	random forests, data imputation (filling missing data)
12 / WEIJIE ZHANG	Bias in random forest variable importance measures: Illustrations, sources and a solution	random forest, variable importance
13	Measuring and Testing Dependence by Correlation of Distances	non-linear correlation
14 /ULISSE GOMARASCA	Visualizing Data using t-SNE	dimensionality reduction, multivariate data visualization
15	The energy of data	non-parametric statistics based on distances
16 / QIAN ZHANG	Summarizing multiple aspects of model performance in a single diagram	model evaluation
17 / WANTONG LI	Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling	model evaluation and calibration
18 / HOONTAEK LEE	Isolation Forest	random forest, novelty/outlier detection
19 / YUNPENG LUO	Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy	uncertainty
20 / SIYUAN WANG	A comparison of techniques for the estimation of model prediction uncertainty	uncertainty
21 /	Verification, validation, and confirmation of numerical models in the earth sciences	model evaluation and calibration
22 / ALBRECHT SCHALL	Hierarchical Clustering via Joint Between-Within Distances: Extending Ward's Minimum Variance Method	clustering
23	Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting	smoothing
24 / JASPER DENISSEN	Toward the true near‐surface wind speed: Error modeling and calibration using triple collocation	uncertainty

2. Feedback

Your feedback is valuable because it helps the instructors and organizers to improve the individual modules and the general structure of the workshop.
The survey results are available here. Statistics and statements should not be taken as an exhaustive or exclusive list.