A very interesting session on #collaborative #learning at #SaTML2024 .

Thanks to all the authors for the exciting insights and to session chair Adam Dziedzic!

Want to know what's happening at SaTML Conference today? Luckily we have Gautam Kamath of University of Waterloo on the ground reporting for us!

First up: this morning's session on privacy chaired by @fraboeni feat. work by Ferry Julien,Junyuan Hong, and others. A thread... 🧵 #SaTML2024 (1/5)

I am very excited to announce that my seminar on #DifferentialPrivacy and #MachineLearning is one of the three seminars nominated for the 'Busy Beaver #Teaching #Award ' (winter'24), the award for 'outstanding and excellent lectures' at the CS department of Saarland University.

Congratulations to @fraboeni Franziska Boenisch for the Fraunhofer Dissertation Award!

Happy to share your recently accepted #ICLR2024 papers on #arxiv ? But spending hours and hours on cleaning off comments, compiling, re-compiling?

You don't have to - just follow the simple guide we compiled for our #PhD #students 👇

sprintml.com/phdlife/2023-0…

Adam Dziedzic and @fraboeni are shortly joining the panel on backdoors in #ML at the BUGS workshop NeurIPS Conference - pass by in room 205 (More info: neurips2023-bugs.github.io) 🎤🎤

Come to #NeurIPS2023 poster 1609 to learn about differentially private prompt learning!

Haonan Duan @fraboeni Adam Dziedzic are doing a great job presenting their work!

Paper: arxiv.org/abs/2305.15594

Visit us this afternoon at 5PM at poster #1619 at NeurIPS Conference to talk about #privacy in #LLM prompting!

🔓Eager for some Privacy-Preserving Machine Learning before Christmas? 🎄
🎓Join us on Dec. 21 for double research seminar with @fraboeni and Adam Dziedzic of CISPA SprintML
ideas-ncbr.pl/en/events/mach…

Visit us this afternoon NeurIPS Conference at poster #1614 where we'll be presenting our work on individualized #privacy in #ML !

Visit our poster currently displayed NeurIPS Conference #1617. Excited to discuss with you about actively defending SSL models.

Machine learning expert Dr. Franziska Boenisch @fraboeni became a CISPA Faculty in September. Her goal is to shape machine learning in such a way that it benefits society while protecting privacy at the same time. Read more here: cispa.de/en/newfaculty_…

I'm excited to present our recent #research on #ML #privacy at the MLSec Machine Learning Seminars. If you're interested, you can still sign up: pralab.github.io/mlsec/

Feeling like you're rushing a submission on trust in ML ICLR 2024 next Thursday? Why not submit to SaTML Conference in a few weeks?

Last year, both authors and reviewers noted that SaTML reviews provided detailed feedback thanks to significant expertise in trustworthy ML on the PC

Looking forward to working with you on exciting topics around #LLMs and #privacy , #trustworthy #ML , and federated learning #FL . Just drop me a message

Testing Robustness Against Unforeseen Adversaries
Max Kaufmann, Daniel Kang (Daniel Kang), Yi Sun, Steven Basart, Xuwang Yin, Mantas Mazeika, Akul Arora, Adam Dziedzic, Franziska Boenisch (@fraboeni), Tom Brown (@nottombrown), Jacob Steinhardt (@JacobSteinhardt , Dan Hendrycks…