Английская Википедия:European Laboratory for Learning and Intelligent Systems

Содержание

Текст

Текст

Текст

Текст

Текст

Текст

Текст

Текст

Текст

ELLIS - the European Laboratory for Learning and Intelligent Systems - is a pan-European AI network of excellence which focuses on fundamental science, technical innovation and societal impact. Founded in 2018, ELLIS builds upon machine learning as the driver for modern AI and aims to secure Europe’s sovereignty in this competitive field by creating a multi-centric AI research laboratory. ELLIS wants to ensure that the highest level of AI research is performed in the open societies of Europe inclduing 39 research units, 14 research programs and a PhD & Postdoc Program.

History

The organization was inspired by the Learning in Machines and Brains program of the Canadian Institute for Advanced Research. ELLIS was first proposed in an open letter to European governments in April 2018, which stated that Europe was not keeping up with the US and China. It urged that European governments act to provide opportunities and funding for world-class AI research in Europe.

It was founded on 6 December 2018 at the Conference on Neural Information Processing Systems (NeurIPS).

Board

The members of the board are:

• Serge Belongie (University of Copenhagen & Cornell University)
• Nicolò Cesa-Bianchi (Università degli Studi di Milano)
• Florence d'Alché-Buc (Télécom Paris)
• Nada Lavrač (Jožef Stefan Institute)
• Neil D. Lawrence (University of Cambridge)
• Nuria Oliver (ELLIS Alicante Unit Foundation | Institute of Humanity-centric AI)
• Bernhard Schölkopf (Max Planck Institute for Intelligent Systems) Chairman
• Josef Sivic (Czech Technical University, École Normale Supérieure & INRIA)
• Sepp Hochreiter (Johannes Kepler University Linz) Board guest

ELLIS units

ELLIS is creating a network of research sites distributed across Europe and Israel. Currently, there are 39 ELLIS units in 14 countries. The long-term goal is to establish a set of world-class ELLIS institutes, each acting as the core of a local AI ecosystem.

ELLIS units

Name	Location	Director(s)	References
ELLIS unit Alicante	Alicante, Spain	Nuria Oliver
ELLIS unit Amsterdam	Amsterdam, Netherlands	Cees Snoek
ELLIS unit Barcelona	Barcelona, Spain	Carme Torras, Dimosthenis Karatzas
ELLIS unit Berlin	Berlin, Germany	Klaus-Robert Müller
ELLIS unit Cambridge	Cambridge, United Kingdom	Carl Edward Rasmussen, José Miguel Hernández-Lobato
ELLIS unit Copenhagen	Copenhagen, Denmark	Ole Winther
ELLIS unit Darmstadt	Darmstadt, Germany	Stefan Roth
ELLIS unit Delft	Delft, Netherlands	Frans Oliehoek
ELLIS unit Edinburgh	Edinburgh, United Kingdom	Ivan Titov
ELLIS unit Freiburg	Freiburg, Germany	Frank Hutter
ELLIS unit Genoa	Genoa, Italy	Lorenzo Rosasco, Massimiliano Pontil
ELLIS unit Graz	Graz, Austria	Wolfgang Maass
ELLIS unit Haifa (Technion)	Haifa, Israel	Shie Mannor
ELLIS unit Heidelberg	Heidelberg, Germany	Oliver Stegle
ELLIS unit Helsinki	Helsinki, Finland	Samuel Kaski
ELLIS unit Jena	Jena, Germany	Markus Reichstein, Joachim Denzler
ELLIS unit Lausanne (EPFL)	Lausanne, Switzerland	Pascal Frossard
ELLIS unit Leuven	Leuven, Belgium	Matthew Blaschko
ELLIS unit Linz	Linz, Austria	Sepp Hochreiter
ELLIS unit Lisbon	Lisbon, Portugal	Mario A. T. Figueiredo
ELLIS unit London (UCL)	London, United Kingdom	Arthur Gretton
ELLIS unit Madrid	Madrid, Spain	Pedro Larrañaga
ELLIS unit Manchester	Manchester, United Kingdom	Magnus Rattray
ELLIS unit Milan	Milan, Italy	Nicolò Cesa-Bianchi
ELLIS unit Modena (Unimore)	Modena, Italy	Rita Cucchiara
ELLIS unit Munich	Munich, Germany	Daniel Cremers, Fabian Theis
ELLIS unit Nijmegen	Nijmegen, Netherlands	Marcel van Gerven
ELLIS unit Oxford	Oxford, United Kingdom	Michael Wooldridge, Stephen Roberts, Yee Whye Teh
ELLIS unit Paris	Paris, France	Gabriel Peyré
ELLIS unit Prague	Prague, Czech Republic	Josef Sivic
ELLIS unit Potsdam	Potsdam, Germany	Ralf Herbrich, Sebastian Reich, Katharina Ladewig
ELLIS unit Saarbrücken	Saarbrücken, Germany	Bernt Schiele
ELLIS unit Sofia	Sofia, Bulgaria	Luc Van Gool
ELLIS unit Stuttgart	Stuttgart, Germany	Andreas Bulling
ELLIS unit Tel Aviv	Tel Aviv, Israel	Amir Globerson
ELLIS unit Trento	Trento, Italy	Nicu Sebe, Bruno Lepri
ELLIS unit Tübingen	Tübingen, Germany	Bernhard Schölkopf, Matthias Bethge
ELLIS unit Turin	Turin, Italy	Tatiana Tommasi
ELLIS unit Vienna (IST Austria)	Vienna, Austria	Christoph Lampert
ELLIS unit Zürich (ETH)	Zürich, Switzerland	Andreas Krause

Research programs

In addition to ELLIS units, a European network of top researchers working at locations throughout Europe is being established, organized into 14 ELLIS Programs. These Programs, directed by outstanding European researchers and including leading researchers as Program Fellows, focus on high-impact problem areas that have the potential to move the needle in modern AI.

The ELLIS Programs are inspired by the CIFAR Program model, and closely collaborate with the CIFAR LMB (Learning in Machines and Brains) Program. Each Program has a budget for 2-3 workshops/year to enable meetings of 10-15 Fellows plus guests for intensive scientific exchange. Workshops can be co-located with academic meetings (usually) held in Europe, or organised as stand-alone events, usually at (and with the support of) ELLIS unit sites.

ELLIS Research Programs

Name	Director(s)
ELLIS Health	Gunnar Rätsch, Oliver Stegle, Mihaela van der Schaar
ELLIS Robot Learning: Closing the Reality Gap!	Tamim Asfour, Aude Billard, Jan Peters
Geometric Deep Learning	Michael Bronstein, Taco Cohen, Max Welling
Human-centric Machine Learning	Plamen P. Angelov, Nuria Oliver, Adrian Weller
Interactive Learning and Interventional Representations	Nicolò Cesa-Bianchi, Andreas Krause, Bernhard Schölkopf
Machine Learning and Computer Vision	Bernt Schiele, Cordelia Schmid, Yair Weiss
Machine Learning for Earth and Climate Sciences	Gustau Camps-Valls, Markus Reichstein
Multimodal Learning Systems	Cees Snoek, Nicu Sebe
Natural Intelligence	Matthias Bethge, Y-Lan Boureau, Peter Dayan
Natural Language Processing	André Martins, Iryna Gurevych, Ivan Titov
Quantum and Physics Based Machine Learning	Hilbert Johan Kappen, Riccardo Zecchina
Robust Machine Learning	Chris Holmes, Samuel Kaski, Yee Whye Teh
Semantic, Symbolic and Interpretable Machine Learning	Volker Tresp, Kristian Kersting, Paolo Frasconi
Theory, Algorithms and Computations of Modern Learning Systems	Francis Bach, Philipp Hennig, Lorenzo Rosasco

PhD & Postdoc Program

The ELLIS PhD & Postdoc Program supports excellent young researchers by connecting them to leading researchers across Europe and offering a variety of networking and training activities, including summer schools and workshops. ELLIS PhDs and postdocs conduct cutting-edge curiosity-driven research in machine learning or a related research area with the goal of publishing in top-tier conferences in the field.

There are two tracks within the ELLIS PhD & Postdoc Program: the academic track and the industry track. These tracks have separate requirements for admission and criteria for activity during the appointment, but otherwise offer the same benefits, network and resources to the applicant.

Academic track

PhD students and postdocs in the academic track strive for international collaboration as they partner with two European academic institutions in their research. These candidates are supervised by two advisors from different European countries: one of the advisors must be an ELLIS fellow/scholar or unit faculty; the other advisor should be an ELLIS fellow, scholar, unit faculty or member. PhDs and postdocs in the program visit the exchange institution for min. 6 months (the partitioning of this time is flexible). Normally, the exchange is partially sponsored by the exchange institution; students can also apply for the ELISE mobility grant to support their exchange.

Industry track

The industry track is open to PhD and postdoc candidates that will be part of a collaboration between an academic institution and an industry partner, and will spend time conducting research at the industry partner during their PhD or postdoc. The candidate will spend a minimum of 50% of their time at the academic institution, and at least 6 months (cumulative) with the industry partner.

For this track, both advisors may be located in the same country. One advisor will represent the academic institution and the other the industry partner. One advisor must be an ELLIS fellow/scholar or unit faculty. The second advisor can be an ELLIS fellow, scholar, member, unit faculty OR it is also possible for the second advisor to join if they have co-authored a paper with the applying student as a result of their collaboration, in a top-tier venue in machine learning or related sub-fields (details here).

Calls and admissions: There is a yearly central call for applications in fall, which will be announced on the ELLIS website. Admission to the program is competitive. For reference, about 5-10% of eligible applicants are accepted via the central call. The next chance to apply via the central recruiting is in fall 2023. Nomination to the program (for students who already started their PhD with ELLIS faculty) is possible year-round.

Timeline for the central recruitment:

• October 2023: application portal opens
• November 15 2023 (23:59 CET): application deadline
• November/December 2023: review phase
• January/February 2024: interviews
• Late February/March 2024: decisions
• Program start: there is no common start for the PhD (depends on the advisor/institution)