The 2nd edition of SaTML Conference is a wrap! It was an absolute honour to co-chair the conference with Carmela Troncoso !

We are very excited to announce the co-chairs for the 3rd edition in 2025: Somesh Jha and Konrad Rieck 🌈

Follow SaTML Conference for updates about the conference!

A research community is only as strong as its members. That's why #SaTML2024 's indebted to our PC, esp these ten members who went beyond the call of duty:
Asia Biega
Konrad Rieck 🌈
Giovanni Cherubin
Jamie Hayes
David Evans
Lydia Zakynthinou
Alina Oprea
David Krueger
Ryan Sheatsley
@fraboeni

Dear all, ever wondered which AI security incidents occur in practice, and how many could be prevented using best practices? Our recent work published in the IAAI incident track covers exactly this! A 🧵1/8

ojs.aaai.org/index.php/AAAI…

We finally released the data from our work on evaluating and calibrating AI models with uncertain ground truth. This is an excellent and realistic new benchmark for machine learning with annotator disagreement - a short thread 🧵:

github.com/google-deepmin…

I am looking for a PostDoc to join my team, to work on robust explanations for malware detection. More information in the job ad. Feel free to reach out, and spread the word! Deadline: April 14, 2024 CS Faculty Jobs

My PhD advisor, Fred Piper, has died. Fred was a mentor, a colleague, and a friend. Like many, he guided my career - in ways both visible and invisible. Let us be inspired by his example: it is possible to be highly successful and remain kind to everyone around you. 🙏

Just one month left before SaTML Conference April 9-11 in Toronto! I am excited to hear from Somesh Jha Deb Raji Yves-A. de Montjoye Sheila McIlraith, as well as the authors of accepted papers, and the competition organizing teams!

There's still time to register! satml.org

The submission deadline is extended! You have almost 20 days to send your work on malware analysis and detection, generative AI or adversarial robustness, and join us in July in Vienna, Austria ☕️

🚀Last-minute check-in to an academic career?

We're seeking two PhD students to research the security and privacy of machine learning. Fully funded positions in vibrant Berlin!

Deadline is this Friday, March 8th. Apply now: mlsec.org/jobs.html#jobs BIFOLD TU Berlin

First, Konrad Rieck 🌈 is a pioneer of ML security and has an awesome lab. Second, Berlin is not too bad to live in. Go apply!

Models often fail under distribution shifts—can pre-training on a large and diverse dataset and then fine-tuning on a task-specific dataset help?
W/ Ben Cohen-Wang, Josh Vendrow we show that this depends on the specific failure mode. In particular, pre-training can help with

Submission link for WoRMA: Workshop on Rethinking Malware Analysis is now online, as well as an update list of PC member and preliminary program schedule! This year I am co-chairing with \/era🌻 and Savino Dambra . Submission deadline approaching: March 14, 2024.

worma.gitlab.io/2024/

Excited to share our work on analysing the resilience of DP-SGD against _specific_ threat models via an information theoretic approach!
arxiv.org/pdf/2402.14397…
w/ Boris Köpf Andrew Paverd Shruti Tople Lukas Wutschitz & Santiago Zanella-Beguelin
To appear at this year's USENIX Security

Do you want me to spend my time reading through 2000+ pages of job applications? If so, apply to come and work with us! (Genuinely, really, please do apply — we’re hiring for 12 new academic posts and it’s a fantastic place to work.)

The 3rd Workshop on Rethinking Malware Analysis (WoRMA) will be co-located with EuroS&P 2024, in Vienna, Austria ☕️🏛️🏔️. Submit your work, be that novel technical research or a position/open-problem paper! worma.gitlab.io/2024/

We'll also have a great keynote by Maura Pintor and a malware analysis demo with Simone Aonzo . More details to follow. We are putting the hotcrp submission link asap.

We open sourced Magika, a new deep learning-based CLI tool/API to do filetype identification. It's fast even when run on a single CPU and outperforms existing approaches. Get it with `pip install magika` and send your feedback to github.com/google/magika!

If you are working at the intersection of Security and GenAI, please consider submitting to this workshop. I expect there is a lot of really exciting early work happening in this area and we'd love to see your paper!

sites.google.com/view/sagai2024…