We have put a new paper called “Single-model attribution via final-layer inversion” about deep generative models on arXiv. Wonderful job, Mike, Jonas, and Asja! 🥁🎷🪗🎻
High-Dimensional Extremes
Together with our amazing collaborator Marco Oesting, we have put a new paper with the title “Extremes in high dimensions: methods and scalable algorithms” on arXiv. Thanks for the great work, Marco! 🍻
Lag selection and stability in AR processes
With Somnath and our collaborator Rainer von Sachs, we have put a new paper with the title “Lag selection and estimation of stable parameters for multiple autoregressive processes through convex programming” on arXiv. Well done, Somnath! 🥇
Mahsa passed final exams
Mahsa passed her doctoral exam, earning her doctorate in Mathematics from Ruhr-University Bochum. Great job, Dr. Taheri! 🎓
Paper on Targeted Deep Learning Accepted
Shih-Ting’s paper on targeted deep learning has been accepted at Stat. Good job, Shih-Ting! ⛰️
DeepMom paper published
Shih-Ting’s paper “DeepMoM: Robust Deep Learning With Median-of-Means” is now published in volume 32 of the Journal of Computational and Graphical Statistics. Find the paper here.
Two new papers on deep learning
VIP Visitor
Professor Yuting Wei will visit us as a Visiting International Professor (VIP), a fellowship awarded by RUB to outstanding international researchers. Thank you, RUB! Congratulations, Yuting: well deserved! And congratulations to our PhD students, who helped make this possible! 🎉🎉
Electronic Journal of Statistics
Johannes has joined the editorial board of the Electronic Journal of Statistics. Find the journal’s website here.
Contributions to the GPSD 2023
We have four short lectures at the 16th German Probability and Statistics Days: The DeepCAR method: forecasting time-series data that have change points (presenter: Ayla), Lag selection and estimation of stable parameters for multiple autoregressive processes through convex programming (Somnath), Reducing computational and statistical complexity in machine learning through cardinality sparsity (Ali), and Extremes in high dimensions: statistical theories and scalable algorithms (M. Oesting, Stuttgart). Johannes co-organizes the sessions on Computational and high-dimensional statistics (with D. Rudolf, Passau). Looking forward to the conference!
Johannes Lederer 