First Machine Learning Gravitational-Wave Search (Mock Data) Challenge – MLGWSC-1 – co-organized by “Werkstatt” members
The first machine learning gravitational-wave search mock data challenge is hosted by the Albert-Einstein-Institut Hannover and the Friedrich-Schiller-Universität Jena (FSU). The organization team at FSU are Ondřej Zelenka, PhD student in “Werkstatt” pilot project P6 “Gravitational Waves”, and the pilot project PI Prof. Dr. Bernd Brügmann.
In this challenge, participants are tasked with finding gravitational-wave signals of varying complexity in a noisy background. Entries are evaluated on metrics which are used to classify the performance of real-world, state-of-the-art search algorithms. The goal of the challenge is to create a collaborative publication that collects state-of-the-art machine learning based gravitational-wave search algorithms and enables a comparison to classical approaches such as matched filtering or coherent burst searches.
More details on the challenge can be found on the github-page and the accompanying wiki. The virtual Kick-off meeting takes place on October 12th, 2021 at 3pm CEST. Interest to participate in the challenge can be expressed via e-mail by December 31st, 2021.
Seminar Series – 2021 Summer Semester
The PhD students of the project present their work.
The seminar series will be continued in the summer semester 2021. Probably again online. Stay tuned.
Seminar Series – 2020/21 Winter Semester
The PhD students of the project present their work.
Wednesdays, 2:00 p.m. to 3:00 p.m
Online event: To receive the login details please contact Dr. Bettina Färber (research coordinator).
03.02.2021
Kevin Lamkiewicz: Reducing Haystacks to Needles: Clustering for viral genome data
Affiliation: Bioinformatics / High-Throughput Analysis
Project: Data-driven virus diagnostics at multiple levels II (Application)
10.02.2021
Ondřej Zelenka: Neural Networks for Gravitational Wave data analysis
Affiliation: Institute for Theoretical Physics
Project: Deep learning for data analysis in gravitational wave astronomy
Wasim Ahmad: Cause-Effect Analysis of Multivariate Time Series using Deep learning
Affiliation: Computer Vision Group
Project: Detection of causal relationships using deep learning
17.02.2021
Nora Abdelmageed: JenTab: Tabular Data to Knowledge Graph Matching
Affiliation: Heinz Nixdorf Chair for Distributed Information Systems
Project: Learning of data annotations
Alina Lopatina: Identification of disease-specific pattern using generative adversarial network
Affiliation: Medical Physics Group at University Hospital
Project: Use and reuse of MRI data in biomedical research
24.02.2021
Andreas Goral: Probabilistic Inference in Conditional Gaussian Models
Affiliation: Theoretical Informatics Chair
Project: Model learning – Probabilistic programming
David Pertzborn: Deep Learning for Maldi-Imaging
Affiliation: Ear Nose and Throat Department at University Hospital
Project: Combined analysis of image data from head and neck cancer
03.03.2021
Matthias Körschens: Self-Supervised Deocclusion for Plant Cover Determination
Affiliation: Institute of Ecology and Evolution
Project: Development, digitization and establishment of sensor-based phenological observations
Henrik Voigt: Introductory Talk: Teaching Agents
Affiliation: Visualization and Explorative Data Analysis Group
Project: A generic conversational interface for scienti c data visualization
Jakob Wolff: Introductory Talk: Connect and learn new materials for optoelectronics
Affiliation: Institute of Condensed Matter Theory and Optics, Botti group
Project: Machine learning optoelectronic properties of materials
Jihen Amara: Introductory Talk: Integrating Knowledge Graphs for DL Interpretability
Associated application project: Integrating Knowledge Graphs for DL Interpretability
New “Werkstatt” Pilot Projects
03.12.2020
The “Werkstatt” project was designed in a way that after two years five pilot projects from application fields (with five new PhD students) will join. After a call in May 2020 with 13 excellent applications, the MSCJ Board of Directors has now selected the five new projects.
In addition, the Board decided to grant two projects, that were very close to the five selected projects, a start-up funding from MSCJ budget to give them time to find an alternative funding. These two projects will be closely associated to the Werkstatt.
Some of the new PhD students already started in October and November. In fact, some already attended the Summer School in September.
Welcome to the project!
Read the complete blog post.
“Werkstatt” Summer School 2020
“Machine Learning for Life and Natural Sciences”
07.-11.09.2020
Workshops:
- Machine Learning (ML), organized by J.-Prof. Dr. Sina Zarrieß
- Deep Learning (DL), organized by Prof. Dr. Joachim Denzler
- Probabilistic Graphical Models (PGM), organized by Prof. Dr. Joachim Giesen
- High-Performance Computing (HPC), organized by Dr. André Sternbeck
- Wednesday: Invited Talk by Dr. Timo Stich (Carl Zeiss AG)
- Hackathon, organized by Dr. Erik Rodner (Carl Zeiss AG)
Information on the schedule, venue and Covid-19 regulations can be found on this website.
Blog post about the Summer School.
Seminar Series – 2020 Summer Semester
The PhD students of the project present their work.
Wednesdays, 2:00 p.m. to 3:00 p.m
Online event: To receive the login details please contact Dr. Bettina Färber (research coordinator).
20.05.2020
Kevin Lamkiewicz: Multiple sequence alignments of Coronavirus
Affiliation: Bioinformatics / High-Throughput Analysis
Project: Application: Coronaviruses
Stephan Peter: Distributed organizations
Affiliation: Biosystems Analysis
Project: Data-driven virus diagnostics at multiple levels I (Methods)
27.05.2020
Ondřej Zelenka: Deep learning for gravitational wave data analysis
Affiliation: Institute for Theoretical Physics
Project: Deep learning for data analysis in gravitational wave astronomy
Matthias Körschens: Towards automatic analysis of plant phenology
Affiliation: Institute of Ecology and Evolution
Project: Development, digitization and establishment of sensor-based phenological observations
03.06.2020
Andreas Goral: Making probabilistic modelling feasible
Affiliation: Theoretical Informatics
Project: Probabilistic Programming
Alina Lopatina: Relevance analysis of deep neural network for classifying multiple sclerosis patients and healthy volunteers based on MR-SWI data
Affiliation: Medical Physics Group at University Hospital
Project: Use and reuse of MRI data in biomedical research
10.06.2020
Nora Abdelmageed: Towards transforming tabular datasets into knowledge graphs
Affiliation: Heinz Nixdorf Chair for Distributed Information Systems
Project: Learning of data annotations
David Pertzborn: Towards multi-modal and 3D image registration
Affiliation: Ear Nose and Throat Department at University Hospital
Project: Combined analysis of image data from head and neck cancer
Wasim Ahmad: Introductory talk
Affiliation: Computer Vision Group
Project: Detection of causal relationships using deep learning