🚀 DL Math&Efficiency Reading Group - Upcoming Talk! Join us for the next session of the DL Math&Efficiency Reading Group, where we explore the theory and practice of efficient and scalable machine learning. 📅 Monday, 20 October 2025 ⏰ 5pm CEST 🎙 Giulia Lanzillotta (ETH AI Center, ETH Zurich, Switzerland) 📝 Topic: Testing Knowledge Distillation Theories with Dataset Size 🔍 Why does knowledge distillation work? And what role does dataset size play? Giulia Lanzillotta explores this overlooked dimension by showing that the test error gap between distillation and standard training widens as dataset size decreases, revealing the data efficiency of distillation. Her results challenge the label smoothing hypothesis and strengthen support for the dark knowledge view, offering fresh insights into the mechanisms behind distillation. 🔗 OpenReview: https://lnkd.in/dQuXYSHX ✨ 🔗 DL Math&Efficiency website: https://lnkd.in/gYzuyDcG ✨ #DLMath&Efficiency #EfficientML
ELLIS Unit Graz
Forschungsdienstleistungen
High-level fundamental machine learning research at Graz University of Technology.
Info
At the ELLIS Unit Graz, located at Graz University of Technology, we focus on high-level fundamental machine learning research and contribute to the fields of Brain-Inspired Machine Learning, Resource-efficient Deep Learning, Human-Centred Computing, Trustworthy AI, Visual Computing, and Probabilistic Machine Learning. The Ellis Unit Graz has strong links with other scientific disciplines through the tight coupling with the Graz Center for Machine Learning (GraML) and collaborations with various industry partners.
- Website
-
https://ellis-unit-graz.github.io/
Externer Link zu ELLIS Unit Graz
- Branche
- Forschungsdienstleistungen
- Größe
- 11–50 Beschäftigte
- Hauptsitz
- Graz
- Art
- Nonprofit
- Spezialgebiete
- Brain-inspired Machine Learning, Resource-efficient Deep Learning, Human-Centred Computing, Trustworthy AI, Visual Computing und Probabilistic Machine Learning
Orte
-
Primär
Wegbeschreibung
Graz, AT
Updates
-
🌱 How can we make AI more energy-efficient and sustainable? Join our ELLIS UnConference Workshop on **Energy-Efficient AI: Models, Algorithms, and Hardware for Sustainable Intelligence** taking place in Copenhagen 🇩🇰 on December 2, 2025 at the Bella Center, co-located with #EurIPS. As artificial intelligence systems grow in scale and capability, their energy demands are increasing at an unsustainable pace. This workshop addresses the urgent need for energy-efficient AI, a field that seeks to develop environmentally responsible and computationally efficient approaches to machine learning. The goal is to explore innovations that reduce energy consumption across the AI stack — from models and learning algorithms to hardware implementations — while maintaining or even enhancing performance. One source of inspiration for reaching this goal is the brain, which manages to provide intelligence with 20W. This workshop aims to bring together researchers and practitioners from diverse backgrounds to foster collaboration and cross-pollination of ideas in making AI more sustainable without compromising its transformative potential. Keynote Speakers: 🔹Angeliki Pantazi (IBM Zurich) 🔹Emre Neftci (Forschungszentrum Jülich) 🔹Wilfred van der Wiel (University of Twente) 🔹Shih-Chii Liu (ETH and University of Zurich) 🔹Iason Chalas (ETH and IBM Zurich) 🔹Nasir Ahmad (Radboud University Nijmegen) 🔹Yukun Yang (Graz University of Technology) Organisers: 🔸 Wolfgang Maass (Graz University of Technology) 🔸 Marcel van Gerven (Radboud University Nijmegen) 📣 Call for contributions open now! 🕒 Submission deadline: October 23, 2025 (AoE) (extended) 🔗 Learn more at eurips.cc/ellis and https://lnkd.in/e948epvH ELLIS Unit Graz ELLIS unit Nijmegen Graz University of Technology Radboud University Nijmegen ELLIS - European Laboratory for Learning and Intelligent Systems
-
ELLIS Unit Graz hat dies direkt geteilt
🎓 The application portal for the #ELLISPhD Program is open today! Info & link: https://bit.ly/45DSe75
-
Thank you ELSA – European Lighthouse on Secure and Safe AI for the warm welcome! We are excited to be part of the network!
Welcome new ELSA members!🎉 Our network is growing, and we are proud to announce that 1️⃣5️⃣ fifteen members have joined ELSA: - Czech Technical University in Prague - ELLIS Unit Delft - Fundación de la Comunitat Valenciana Unidad (ELLIS Alicante Foundation) - Friedrich Schiller University Jena - Technische Universität Graz - IDEAS Research Institute - Institute for Machine Learning, Johannes Kepler Universität Linz - Institute for Systems and Robotics (ISR-Lisboa) - Institute of Science and Technology Austria - INESC-ID (Instituto de Engenharia de Sistemas e Computadores – Instituto de Desenvolvimento) - Instituto de Telecomunicações - Instituto Superior Técnico - Max Planck Institute for Biogeochemistry - Ukrainian Catholic University - Albert-Ludwigs-Universität Freiburg im Breisgau Welcome! Incorporating these new members will enable ELSA to expand its reach across Europe and strengthen and intensify research connections. We are excited about enhancing ELSA and facilitating connections between researchers across Europe. You can find all the info on our website. Also, look at our network page, where you can find the new members as well as the contact persons. https://lnkd.in/ehRV5Cgn #ELSAAI #network #expansion Mario Fritz ELLIS - European Laboratory for Learning and Intelligent Systems
-
-
The ELLIS Unit Graz has joined LinkedIn! Stay tuned for news from and events of our growing ELLIS Community at Graz University of Technology! Wolfgang Maass Robert Legenstein Elisabeth Lex Robert Peharz Ozan Özdenizci Olga Saukh Thomas Pock Franz Pernkopf Horst Bischof Bernhard Geiger