🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Big congratulations Arno, richly deserved!

I want to highlight another item in addition to Arno who is clearly a rising star if not a star already: A future rising star Daolang Huang was a runner-up in the race. We will certainly hear more of him later! (Sorry to induce pressure on you Daolang...)

The new ELLIS Institute is excellent news for Europe and Finland. A significant investment to the future of Europe - its growth, and security of supply of expertise. Visionary act by Finnish Government, Ministry of Education and Culture, and Peter Sarlin!

Exciting news! Finland to establish the 2nd ELLIS Institute, supported by gov't & industry. 🇫🇮Coordinated by Finnish Center for AI 🦣 @[email protected], Aalto University & ELLIS Unit Helsinki, this initiative strengthens Europe's AI ecosystem. #ELLISforEurope bit.ly/3YuuIoa

This time I'll be recruiting an outstanding PhD student, to be co-supervised with @SamiKaski, for a cool project on learning theory and methods for novel types of distributional shifts. The successful candidate will be based at Computer Science @ The University of Manchester ✨ Get in touch if interested!

🚨New speaker alert! We're excited to announce Tom Griffiths Griffiths Computational Cognitive Science Lab as one of our keynote speakers ✨ Don't miss this opportunity to connect & share your ideas. 📢 Submissions are open until the 25th of November! 🌐 sites.google.com/view/caihu25/c…

Optimizing decision utility in Bayesian experimental design is key to improving downstream decision-making. Excited to share our #NeurIPS2024 paper on Amortized Decision-Aware Bayesian Experimental Design: arxiv.org/abs/2411.02064 Luigi Acerbi Samuel Kaski Finnish Center for AI 🦣 @[email protected] Details below.

3/ Improving robustness to corruptions with multiplicative weight perturbations (spotlight) led by Trung Trinh and with Markus Heinonen and Samuel Kaski arxiv.org/abs/2406.16540

🌟 Spotlight Paper at #NeurIPS2024! 🌟 Making AI models more robust just got easier! Our paper introduces DAMP - a simple yet powerful method that helps neural networks handle corrupted images better, without hurting performance on clean data.

9/9 📄 Read the full paper and try our code: Paper: arxiv.org/abs/2406.16540 Website: trungtrinh44.github.io/DAMP/ Joint work with Markus Heinonen Luigi Acerbi Samuel Kaski #MachineLearning #ComputerVision #DeepLearning Like & RT to share this research! 🙏

Good morning #neurips2024! We've got a lot of cool work to share with you from 🇫🇮 (Aalto University + University of Helsinki), check it out 👇 fcai.fi/news/2024/fcai… cc Faculty of Science, University of Helsinki HIIT Computer Science Computer Science @ University of Helsinki ELLIS Samuel Kaski

Postdoc, research fellow and PhD positions: the call to join the Finnish Center for #ArtificialIntelligence is now open. Check out our research areas and supervisors and apply by Feb 2, 2025: fcai.fi/winter-2025-re…

"When do GFlowNets Learn the Right Distribution?" is accepted at ICLR 2025! 🎉 We explore:🔹How balance violations impact learned distributions; 🔹 Why some distributions are inherently unattainable;🔹 A new, theoretically grounded assessment metric. openreview.net/forum?id=9GsgC…