Inaugural Lectures - Prof. Anne-Marie Kermarrec, Prof. Lenka Zdeborová, Prof. Edouard Bugnion

Thumbnail

Event details

Date 22.11.2022 17:4520:00  
Speaker Prof. Edouard BugnionProf. Anne-Marie Kermarrec, Prof. Lenka Zdeborová
Location
Category Inaugural lectures - Honorary Lecture
Event Language English
Date: 22 November 2022

Program: 
  • 17:45-17:50: Introduction by Prof. Rüdiger Urbanke, Dean of the IC School
  • 17:50-18:20: Inaugural Lecture Prof. Anne-Marie Kermarrec
  • 18:20-18:25: Q & A
  • 18:25-18:30: Introduction by Prof. Henrik M. Rønnow, Director of IPHYS
  • 18:30-19:00: Inaugural Lecture Prof. Lenka Zdeborová
  • 19:00-19:05: Q & A
  • 19:05-19:10: Introduction by Prof. Rüdiger Urbanke, Dean of the IC School
  • 19:10-19:40: Inaugural Lecture Prof. Edouard Bugnion
  • 19:40-19:45: Q & A
  • 19:45: Apéritif at Piano restaurant, SG Building         
Location:  SG 1

Registration: Click here

***********************************************************
Prof. Anne-Marie Kermarrec

Internet Collaboratif : Retour vers le Futur 

Abstract
Internet a été initialement conçu avec comme objectif de devenir un medium démocratique, permettant aux citoyens du monde d'échanger leurs données sans entrave aucune. L’avènement des géants du Web (les fameux GAFAM et autres BATX)  a  singulièrement remis en cause cette belle idée lorsqu’ils s'en sont emparés pour collecter massivement nos données et nous les revendre sous une forme ou une autre.  Cet exposé prend le parti optimiste défendant l'idée que cet objectif peut être atteint grâce aux avancées réalisées en informatique, et en particulier ce que l'on appelle l'informatique distribuée. J'illustrerai cette thèse à travers un service central d’Internet particulièrement important, celui de la recommandation de contenu et son évolution tant dans le monde académique que dans l’industrie.
 
About the speaker
Anne-Marie Kermarrec est Professeur à l’EPFL depuis 2020. Jusqu’en 2019, elle a dirigé la startup qu’elle a fondée en 2015, Mediego, issue de ses travaux de recherche, qui commercialise des produits de recommandation de contenus à destination de la presse en ligne. Elle l’a cédée en 2019. Auparavant, elle a été chercheuse à l’université d’Amsterdam (1996-97), Maitre de conférences à l’université de Rennes (1998-2000) et chercheuse à Microsoft Research Cambridge, UK (2000-2004) puis directrice de recherches à Inria (2004-2015). Elle a obtenu une ERC en 2008 and une ERC proof of Concept in 2013. Elle a reçu, de l’Académie des Sciences, le Prix  Montpetit en 2011 et le prix de l’Innovation en 2017. Elle est ACM Fellow depuis 2016 et a été décorée du titre de Chevalier de la Légion d’honneur en 2018.

***********************************************************
Prof. Lenka Zdeborová

The Physics of Algorithms

Abstract
Computational questions in high-dimensional problems are ubiquitous, yet we still lack a satisfying theoretical framework able to answer most of them. Corresponding problems sometimes map to systems studied in statistical physics where the high dimension translates into a large number of interacting elements. Theoretical tools of statistical physics can then be deployed to study computational problems and inform both mathematical and algorithmic developments. We will highlight some key results in this field with examples of applications in artificial neural networks and in signal processing.

About the speaker
Lenka Zdeborová received her PhD in physics from University Paris-Sud and from Charles University in Prague in 2008. She spent two years in the Los Alamos National Laboratory as the Director's Postdoctoral Fellow. Between 2010 and 2020 she was a researcher at CNRS working in the Institute of Theoretical Physics in CEA Saclay, France. In 2014, she was awarded the CNRS bronze medal, in 2016 Philippe Meyer prize in theoretical physics, in 2018 the Irène Joliot-Curie prize, in 2021 the Gibbs lectureship of AMS.

***********************************************************
Prof. Edouard Bugnion

At Microsecond-Scale, Inside a Datacenter

Abstract
Our cell phones are the defining, visible innovation of our era, yet datacenters form the invisible part of that digital iceberg. Within a datacenter, thousands or more machines communicate through very short remote procedure calls  that often take a few microseconds to service but must, nevertheless, meet tight service-level objectives expressed in terms of tail latency. At such microsecond-scale, operating systems and network overheads dominate and limit overall system efficiency.

I will present key results from the EPFL Datacenter Systems Laboratory that (1) leverage virtualization hardware for energy proportionality and tail-tolerance, (2) innovate at the network protocol layer to eliminate head-of-line blocking and offer load balancing and consensus within the transport.

About the speaker
Prof. Edouard Bugnion joined EPFL in 2012, with a teaching and research focus on datacenter systems. Before joining EPFL, he spent 18 years in the US, where he received his PhD from Stanford, and co-founded two startups: VMware and Nuova Systems (acquired by Cisco). He served as VMware’s first CTO and was later the VP/CTO of Cisco’s Server, Access, and Virtualization Technology Group. At EPFL, he is the Academic Co-director of the Swiss Data Science Center, he has served as the Vice-President for Information Systems (2017-20) as well as on the COVID-19 Independent Scientific Task Force for the Swiss Confederation (2020-22). Prof. Bugnion is an ACM Fellow and a member of the Swiss Academy of Technical Sciences (SATW). He received the SIGOPS Hall of Fame Award for Disco, the ACM Systems Award for VMware, and Best Paper Awards at SOSP, OSDI and Eurosys. 





 

Practical information

  • Informed public
  • Registration required

Organizer

Tags

inaugural lecture Edouard Bugnion Anne-Marie Kermarrec Lenka Zdeborová

Share