Another glitch in the simulation.

Our Trustworthy Machine Learning textbook is out today! 🥳The book’s dedicated webpage is trustworthyml.io, and you can also find it on arXiv under arxiv.org/abs/2310.08215. Check out this 🧵on what key questions this freely available book answers 👇(1/6)

"We are looking for a motivated individual to fill this postdoc position." Actually, nowadays if someone wants to be a postdoc, they are already very motivated, okay?

Excited to share our latest #NeurIPS2023 paper "Labeling Neural Representations with Inverse Recognition." (arxiv.org/abs/2311.13594) We present INVERT — new global explanation method that explains neurons with human-understandable concepts they detect! A thread 🧵(1/8)

Wishing everyone a joyous holiday season filled with warmth, laughter, and all-around great Christmas vibes! 🎄✨

Enough’s been written about the break-neck pace of publishing in CS. I want to highlight a related & also problematic aspect: citation patterns. Look at where papers are published, which papers they cite, and where those cited papers are published. You’ll find a closed system.

#Brandenburg zeigt Haltung! Die Wissenschaft lebt von Offenheit & Internationalität. Wir, das ATB, stehen für ein weltoffenes, respektvolles & solidarisches Miteinander und stellen uns entschieden gegen Hass und Ausgrenzung. 👉change.org/zeigthaltung 👈 #haltungzeigen

Our paper on dangers of watermark detector neurons in Deep Neural Networks is now published at Springer CCIS! 🎉 If you haven’t seen it — check the thread below! Published article: link.springer.com/chapter/10.100… #ExplainableAI #MachineLearning

ERCIM Fellowship is an amazing opportunity for PhD holders for research mobility and career advancement. It is also an excellent initiative for democratizing research. As an ex fellow, couldn't recommend it more. Deadline is April 30th.

Germany tried this. It is a disaster. Pushes ppl out of research. The positions they are supposed to transition into don't exist. I can't keep my research team here because of arbitrary contract limits like this. Tragic and demoralizing for all of us. #WissZeitVG #IchbinHanna

NIH Post-PhD scientists should be able to get a scientist job like any normal job. In a normal job market, if you are a person with skills and continue to be valuable to your employer, your employer will want to keep you around and pay you commensurate to your value. That's it.

We have two open #PhD positions. It's a great place to do cutting-edge #AI and #bioeconomy research and solve real problems with a chance to make an impact. Also, #Potsdam is beautiful.

Except that they eat more chocolate, I suppose.

📢 Post-Bayesian online seminar series coming!📢 To stay posted, sign up at tinyurl.com/postBayes We'll discuss cutting-edge methods for posteriors that no longer rely on Bayes Theorem. (e.g., PAC-Bayes, generalised Bayes, Martingale posteriors, ...) Pls circulate widely!

Also, people with BAs in philosophy and such are here to enlighten us on why no one needs a PhD in ML.

Aktarmalı yurtdışı uçuşumu 10 saat kala iptal ettiğiniz için toplantılarıma yetişemiyorum. Biletimi bir sonraki güne değiştirdiniz değiştirmesine de, ben bunu iş yerime nasıl açıklayacağım? Üstelik aradaki fiyat farkını da iade etmeyeceğinizi söylüyorsunuz. AJet AJet Destek

Today, we are publishing the first-ever International AI Safety Report, backed by 30 countries and the OECD, UN, and EU. It summarises the state of the science on AI capabilities and risks, and how to mitigate those risks. 🧵 Link to full Report: assets.publishing.service.gov.uk/media/679a0c48… 1/16

We are excited to release a short course on AGI safety! The course offers a concise and accessible introduction to AI alignment problems and our technical & governance approaches, consisting of short recorded talks and exercises (75 minutes total). deepmindsafetyresearch.medium.com/1072adb7912c

This is Booking.com: you can send a message, but don’t expect a response. You can call them 24/7 too… if you can somehow find the number they won’t actually give you. Amazing way to damage your reputation after firing talented people for AI. Well done! 👏

Dear #AI #ComputerVision #AIEthics community, I received a review request for a manuscript on facial ethnicity recognition, and it feels wrong. Does this task have any scientific basis, validity, and value? If not, any references I could cite to justify rejecting it?