ELLIS - European Laboratory for Learning and Intelligent Systems

🏹 Job alert: DTU Compute has an attractive group leader opening as Tenure Track Assistant / Associate Professor in AI. The position is initially funded by the Pioneer Centre for AI. DTU Compute is an active partner in the ELLIS Unit Copenhagen. Excellent candidates in any field of machine learning theory, algorithms and computation or in AI applications in e.g., the social, neuro, or natural sciences, are encouraged to apply. The position comes with a start package funded by the Pioneer Centre. The package will allow you to establish a group of junior scientists. 📍 Kgs. Lyngby near Copenhagen 📅 Application deadline: April 30 🔗 More info: https://lnkd.in/dQxZ4hVh

💬 "Teaching language models to speak chemistry: From design to synthesis" is the focus of today’s AI for Good webinar, presented by Philippe Schwaller (EPFL). ⚡ Philippe Schwaller (ELLIS Scholar at Unit Lausanne) participates in the ELLIS Program "Machine Learning for Molecule Discovery" and will talk about how Artificial Intelligence is revolutionizing chemical research by bridging computational design and experimental synthesis. ChemCrow, a large language model, can directly control robotic synthesis platforms, successfully creating compounds. Saturn, a smaller language model, optimizes molecular design with synthesizability constraints, ensuring practical accessibility. Together, these AI systems transform chemical research from design to synthesis, with potential impacts on drug discovery, catalysis, and materials development. 📅 Mon, March 31, 2025 🕓 16:00-17:00 CET 🎤 Philippe Schwaller (EPFL) 🔗 Registration: https://lnkd.in/gHyn2Kgu This talk is part of the AI for Good webinar series “From Molecules to Models”, highlighting significant contributions from various ELLIS Programs in the field of AI for life sciences.

🏹 Job alert: The PURRlab (Pattern Recognition Revisited lab) at the IT-Universitetet i København invites highly motivated individuals to apply for 2 year PhD or Postdoc positions on evaluation metrics for multimodal health data. Machine learning competitions drive algorithm development in healthcare but often lack real-world robustness. Relying on a single metric like accuracy leads to overly similar models that fail on diverse patient cases and have high training costs. Such competitions may also discourage underrepresented groups from participating. To address this, the researchers will design competitions with multiple metrics across different patient subgroups, develop methods to assess and enhance algorithm diversity, and explore ways to improve robustness while reducing carbon footprint. They will also study how these changes impact inclusivity in data science. 📍 Copenhagen 📅 Application deadline: April 30 🔗 More info: https://lnkd.in/ejhmkYaR

The ELLIS Unit Potsdam invites you to the 2nd Symposium "AI in Public Health Research" which offers a platform for exchange between AI science and research, public health practice, and policy with the aim of redefining the possibilities and limitations of the application of AI in public health research. You are invited to join the conversation and help shape the future of AI-driven public health research. 📌 Robert Koch Institute in Berlin 📅 May 14-15, 2025 🔗 https://lnkd.in/eJubgj2u Deadline to register is April 25.

🏹 Job alert: The Institute of Computational Biology at Helmholtz Munich is looking for a Postdoctoral Researcher (f/m/x) in Machine Learning or Computational Biology to join the Theis Lab, in collaboration with the Oxford Centre of Neuroinflammation. The position focuses on developing graph-based deep learning and optimal transport methods to analyze brain disease mechanisms using multi-omics data, including spatial transcriptomics and mass spectrometry. 📍 Neuherberg near Munich 📅 Application deadline: April 11 🔗 https://lnkd.in/e6tgwFDN

Say hello to Iryna Gurevych, Professor at Technische Universität Darmstadt 🇩🇪 , MBZUAI (Mohamed bin Zayed University of Artificial Intelligence) 🇦🇪 and INSAIT - Institute for Computer Science, Artificial Intelligence and Technology 🇧🇬 . She's an ELLIS Fellow, Co-Director of the ELLIS Program Natural Language Processing, and a member of the ELLIS Unit Darmstadt. She said that by communicating about her career and the scientific process, she hopes to inspire junior researchers to follow their own path. ✨ ✨ ✨ #WomenInELLIS We're spotlighting women in ELLIS in 2025: celebrating their achievements, leadership, guidance, and innovation in the exciting fields of AI and machine learning across Europe.

🧬 🔭 💊 Last week's Helmholtz-ELLIS Workshop on Foundation Models in Science brought together leading researchers to explore how AI is transforming fields like materials science, genomics, and astronomy. 🤝 Jointly organized by the Helmholtz Foundation Model Initiative and the ELLIS Units in Heidelberg, Munich and Potsdam, the event featured leading experts from academia and industry discussing innovations such as AI-driven material discovery, metagenomic modeling for pandemic tracking, and foundation models for structured scientific data. 💡 Unlike large conferences, this workshop provided an intimate setting for deep, cross-disciplinary discussions, fostering collaboration between academia and industry while addressing key AI challenges such as benchmarking, evaluation, and ethical deployment. 📖 Read the full article to explore the main takeaways from the talks and discussions: https://lnkd.in/eVJBEV_F Helmholtz AI - Helmholtz Imaging - HEIBRiDS - Helmholtz Einstein International Berlin Research School in Data Science - Zuse School ELIZA - Max Delbrück Center

🏹 Job alert: CISPA Helmholtz Center for Information Security is seeking Tenure-Track Faculty in all areas related to Information Security and Artificial Intelligence (f/m/d). CISPA is a world-leading research center that focuses on Information Security and Artificial Intelligence at large. Applicants are expected to grow a research team that pursues an internationally visible research agenda. We provide institutional base funding for three full-time researcher positions and a generous budget for expenditures. 📍 St. Ingbert, near Saarbrücken 📅 Apply by April 8 🔗 More info: https://bit.ly/4i12JV7

💬 Check out the AI for Good webinar on the topic of "Low energy learning from the analysis of learning landscapes in deep and recurrent neural networks" presented by Riccardo Zecchina of Università Bocconi. ⚡ Riccardo is the Co-Director of the ELLIS Program "Quantum and Physics Based Machine Learning" and will speak on investigating non-convex deep and recurrent network landscapes through statistical physics, uncovering key geometric characteristics that inform the design of algorithms which rely on fully local learning schemes. His study examines the coexistence of typical minima, which adhere to the Overlap Gap Property, alongside infrequent flat regions that do not. Then he'll discuss the algorithmic implications for the design of energy preserving local learning in deep architectures. 📅 Mon, March 24, 2025 🕓 16:00-17:00 CET 🎤 Riccardo Zecchina; moderated by Conrad Heinz Philipp of ELLIS unit Jena 🔗 Registration: https://lnkd.in/eNK4gSu7 This talk is part of the AI for Good webinar series “From Molecules to Models”, highlighting significant contributions from various ELLIS Programs in the field of AI for life sciences.

The ELSA – European Lighthouse on Secure and Safe AI Cybersecurity Use Case team announced a new challenge: the "Robust Android Malware Detection Competition" taking place at the 3rd IEEE Computer Society Conference on Secure & Trustworthy Machine Learning Conference on April 9-11 in Copenhagen 🇩🇰. The competition aims to evaluate ML-based detectors concerning ⚡ Adversarial Robustness to Feature-space Attacks ⚡ Adversarial Robustness to Problem-space Attacks ⚡ Temporal Robustness to Data Drift Deadline to register is March 31. Learn more: https://lnkd.in/eVa5D95M