Online Coffee lectures - Overview
The NFDI4Earth Academy
offers its Online Coffee Lecture series
from October 2024 to March 2025. Short lectures and talks on various topics related to Machine Learning, research data and the NFDI will take place every second Tuesday from 11:00 to 12:00 am online via Zoom. This part of the NFDI4Earth Academy program is open to the whole community.
Everybody is welcome and the participation is free of charge. Please register for lectures to receive the dial-in data.
Grab a cup of coffee or tea and drop by virtually!
Online Coffee lecture - Program October 2024 - March 2025
2024
Tuesday 29.10.
11-12 am
Introducing Data Train: Training in Research Data Management and Data Science (Lina Schaare, U Bremen Research Alliance)
For early career researchers, developing strong
skills in data science, research data management, and data literacy is
crucial to their personal and professional growth.
The U Bremen Research Alliance offers the central graduate training program
Data Train as an additional offering for researchers of all disciplines
to enhance their data skills. Data Train is free of costs for participants and offers lectures, workshops and events for interdisciplinary exchange.
Since November 2023, Data Train has been integrated into DataNord, the interdisciplinary data competence center for
the Bremen region, which is associated with consortia of the German
National Research Data Infrastructure (NFDI) initiative.
To meet the high demand for data competencies in academia, industry and society at large,
further development of the Data Train program is supported by the
Federal Ministry of Education and Research and financed by the European
Union – NextGenerationEU.
The talk will give an overview of the Data Train
program and its contents, as well as highlight opportunities for
NFDI4Earth researchers to get involved in the 2025 edition of Data
Train.
Tuesday 12.11.
09-12 am
NLP & LLMs: Basics for Harnessing the power of language (Matthias Aßenmacher, LMU Munich, Munich Center for Machine Learning (MCML), BERD@NFDI)
This 3h online workshop is ideal for researchers who want to learn about the potential of Natural Language Processing (NLP) and Large Language Models (LLMs) for their research and gain a deeper understanding of the inner workings of those models which are frequently perceived as black boxes. You will receive a broad overview of the theoretical foundations of NLP, learn about practical considerations, and get an idea of how to apply this technology to real-world problems. A more exhaustive variant of this course is available at https://slds-lmu.github.io/dl4nlp/ (full lecture). This lecture is in cooperation with the BERD Academy. Register here
Tuesday 26.11. 11-12 am
Introducting the eXplainable Artificial Intelligence (XAI)-Toolbox (Manjunatha Veerappa, IOSB)
The XAI toolbox developed by Fraunhofer IOSB is designed with AI explainability at its core. It enables the swift evaluation of various XAI methods for AI procedures. That is, the toolbox can be used, for example, for data analysis, debugging, and explaining the prediction of any black-box model. This ensures trustworthiness in AI decisions. Currently, it supports scikit and TensorFlow based time series as well as image classifiers. Register here
Tuesday 10.12. 11-12 am
Data Competence Centers: Enhancing Data-Driven Research with Come2Data, DataNord & WiNoDa (Lena Steinmann, Universität Bremen & Data Science Center; Tanja Hörner, Universität Bremen & U Bremen Research Alliance; Alex Miklashevsky, Museum für Naturkunde Berlin)
The German Federal Government’s data strategy and the BMBF’s Action Plan for Research Data aim to strengthen data literacy in science by supporting researchers in managing the growing volume of research data. To foster innovation, the BMBF is funding the establishment of data competence centers at universities and research institutions across Germany. These centers offer comprehensive support for all stages of the data lifecycle, from collection to reuse, and serve as hubs for learning, collaboration, and advancing data-driven research, helping to create a culture of effective data use and analysis in science. NFDI4Earth is collaborating with three centers: Come2Data, WiNoDa and DataNord.
In a Saxon-regional approach, Come2Data (Competence Center for Interdisciplinary Data Sciences) bundles existing data science training and support services as well as expertise and commitment to research data management, NFDI, high-performance computing and analysis methods for data-intensive interdisciplinary research applications such as artificial intelligence and data modeling. DataNord (Interdisciplinary Data Competence Centre for the Bremen Region) offers researchers from all disciplines and career stages a wide range of services to enhance their data handling skills throughout the entire data lifecycle; these include training sessions, hackathons, consulting services, opportunities for networking and professional exchange. The WiNoDa project (Knowledge Lab for Natural Science Collections and Object-Related Data) is creating a data competence center to support researchers in effectively utilizing these collections and advancing research in this domain. Register here
2025
Tuesday 21.01. 11-12 am
tbh
Tuesday 04.02. 11-12 am
Internal Academy Session
Tuesday 18.02. 11-12 am
The OneStop4All - entry point to the resources of the NFDI4Earth (Sibylle Haßler, KIT; Ivonne Anders, DKRZ; Christin Henzen, TUDD)
The OneStop4All (OS4A) is the main access point to all NFDI4Earth resources and aims to support all searches and questions that the users from the Earth System Sciences (ESS) might have. This includes general principles of FAIR data and data management, finding and accessing (primarily ESS relevant) datasets, tools and services as well as training materials, and information about the NFDI4Earth, the involved organisations and participants.
We present the portal and its current functionalities and give some insight into the concept and development. Register here
Tuesday 04.03. 11-12 am
Introducing Galaxy for Earth System (Jérôme Detoc, Ifremer;Marie Jossé, Data Terra (CNRS))
Galaxy now supports Earth sciences, including climate, ecology, and environmental research. With its open-source platform and commitment to FAIR principles, Galaxy provides a powerful set of tools for building, sharing, and replicating workflows. From novice to expert, researchers can access user-friendly tools for comprehensive data analysis without programming skills. Join a growing community leveraging advanced technologies and collaborative resources to drive insights into our planet’s systems—perfect for Earth data enthusiasts looking to elevate their research. Register here
Tuesday 18.3. 11-12 am
tbh
Online Coffee lecture - Program Spring 2024
Friday 12.01. 11-12 am GFZ Data Services (Kirsten Elger, GFZ)
GFZ Data Services is a repository for research data and scientific software across the Earth System Sciences, hosted at GFZ. The curated data are archived, persistently accessible and published with DOI. They range from large dynamic datasets from global monitoring networks with real-time acquisition, to international services in geodesy and geophysics, to the full suite of small and highly heterogeneous datasets collected by individual researchers or small teams ("long-tail data"). In addition to the DOI registration and data archiving itself, GFZ Data Services team offers comprehensive consultation by domain scientists and IT specialists.
This presentation will introduce to the broad service portfolio of GFZ Data Services, including project-specific DOI landing pages for our national and international partners, data curation practices, supporting tools like the online metadata editor, data description templates and extensive data publication guidelines. It will further show examples of how metadata exchange with other data portals is increasing the visibility of our data publications.
Friday 26.01. 11-12 am FID GEO: Library services to make your research more visible and accessible (Melanie Lorenz, GFZ)
Cancelled! Our apologies
The FachInformationsDienst für Geowissenschaften (FID GEO, Specialized Information Service for Geosciences) is a nationwide service for all geoscientists working in Germany. Our team of highly networked librarians, data publishing experts and geoscientists provides services and advice on Open Science issues in the geosciences. We offer Open Access electronic publishing of your research results such as research data, software and texts via our domain repositories (GEO-LEOe-docs and GFZ Data Services) as well as digitisation. This lecture will be recorded for internal usage.
Friday 09.02. 11-12 am
is unfortunately cancelled
Wednesday 21.02. 11-12 am Geo-INQUIRE (Fabrice Cotton, Stefanie Weege, GFZ)
Geo-INQUIRE (Geosphere
INfrastructures for QUestions into Integrated REsearch) fosters
excellent, interdisciplinary and curiosity-driven research of the
solid Earth, including land-sea-atmosphere interfaces. The EU
project benefits from a unique partnership of 51 European
partners. Now, the 1st Geo-INQUIRE Transnational Access Call to European
research infrastructures is open and provides access to unique
high-level installations and experiments. Geo-INQUIRE is welcoming
applications from researchers, technicians, students and other
specialists working in the fields of research covered by the
installations. The call is open until 15 March 2024. In addition, Geo-INQUIRE offers online seminars, online training,
on-site workshops, on-site summer schools and in the near future
personalised training, all free of charge.
We promote equality, diversity and inclusion in all Geo-INQUIRE
activities. Therefore, applications from women and applications
from people working in one of the widening countries, associated
countries and outermost regions of Horizon Europe will be given
special attention during the whole project life time. For
questions feel free to contact geo-i_pmo@gfz-potsdam.de
Friday 08.03. 11-12 am
Inferring Causal Relationships from Data - An Introduction
(Jonas Wahl, DLR & TU Berlin)
One of the primary goals of many research fields is to infer and describe causal relationships between scientific phenomena based on observational and experimental data. In the last decades, scientists and statisticians have developed a variety of formalisms as a scientific language for causal claims, as well as many algorithms that aim to infer causal networks or causal effects under different assumptions. I will give an introduction to the Causal Inference framework of Pearl and others and describe its current state-of-the-art with a particular focus on time series data.
Friday 22.03. 11-12 am Explainable Machine Learning: Basics & Beyond
(Lorenz Linhardt, TU Berlin)
Machine learning (ML) methods have become ubiquitous in the sciences. Despite their widespread usage, it is unclear what they learn and where they may fail. Explainable ML methods attempt to provide additional insights into the inner workings of ML models and their decision-making strategies. This talk will give a brief introduction to explainable ML, in particular well-established methods, and will provide pointers to more recent developments.
Friday 12.04. 11-12 am
The DataHub - empowering digital data-driven research
(Martin Hammitzsch, GFZ)
The DataHub, a collaborative effort across the seven research centers
within the Research Field Earth and Environment of the Helmholtz
Association, aims to break down data silos and empower research. This
integrative initiative, anchored in the field's joint research program,
brings together previously separate marine, terrestrial, and atmospheric
research data into a single, open, and interoperable information hub.
Following the FAIR principles, the DataHub will serve as a digital
ecosystem for researchers to store, manage, share, and analyze their
data across disciplines and institutions. It provides a federated data
infrastructure and a suite of tools for advancing research
digitalization and fostering a cultural shift within the field Earth and
Environment. This interconnected and interoperable data infrastructure
stands as a significant contribution to the National Research Data
Infrastructure (NFDI) - notably, it will support NFDI4Earth.
Friday 26.04. 11-12 am
SIS MAPS (FID Karten) - Specialised Information Service Cartography and Geodata
(Cornelia Koch & Martin Jeske, Staatsbibliothek zu Berlin)
The SIS Maps, located in the Map Department of the Staatsbibliothek zu Berlin (SBB) is funded by the German Research Foundation (DFG). The lecture will focus on the services of the SIS, which are aimed in particular at scientists of geoinformation but also at the interdisciplinary professional community who use maps. They include the largely digital provision of specialist information on cartographic materials (literature, glossary, maps and geodata) as well as infrastructures that enable them to be found and accessed (special catalogues, bibliography and repository).
Online Coffee lecture - Program 2023
Thursday 24.08. 11-12 am Technical issues in data analytics: tips, tricks and hacks (Christoph Lehmann, TU Dresden)
Friday 08.09. 11-12 am Introduction to NFDI4Earth IG High-performance Computing in Earth System Sciences
(Stephan Frickenhaus, AWI) &
IG Long-term Storage and Archiving
(Tim Schürmann, BKG)
The Interest Groups are forum for the NFDI4Earth community to e.g.
identify, discuss and communicate requirements for NFDI4Earth,
jointly advance NFDI4Earth-related concepts, technologies and
processes, implement NFDI4Earth principles, or to disseminate
specific NFDI4Earth offers. Today both interest groups will briefly
introduce themselves, their current work, goals and possibilities to
join the activities.
Friday 22.09. 11-12 am All you need to know about Software Licenses as a RSE (Tobias Schlauch, DLR)
A basic understanding of software
licenses and related topics is necessary for RSEs to successfully
publish their research software from the legal point of view. Typical
questions are: What license should I use for code, documentation, or
data? Can I combine this specific set of software licenses? How to
document this information?
This talk gives an overview about the copyright and software license
basics. Using an example of a data analysis script written in Python, we
go through the process of clarifying all relevant aspects and
documenting them in a practical way. In this context, we introduce REUSE SOFTWARE which provides a standardized approach to document copyright and license information as well as helpful tools.
Friday 20.10. 11-12 am Introduction to PANGAEA (Lars Möller, MARUM)