BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IEM Seminar series: Software engineering for robust Machine-Learni
 ng enabled systems
DTSTART:20220608T150000
DTEND:20220608T160000
DTSTAMP:20260408T021237Z
UID:d6e15c84c1edec4d34c65bec8422de8c269cee5b974806deeae6df8d
CATEGORIES:Conferences - Seminars
DESCRIPTION:Yves Le Traon\nUniversity of Luxembourg\nAbstract:\nNowadays\,
  Machine Learning (ML) provides engineers with the prospect of producing d
 ata-driven software\, with little manual code writing. These ML-enabled so
 ftware bring us to a new era where systems’ logic is automatically produ
 ced from data\, with a small amount of human-written code. \n\nThis radic
 al change questions the way software are engineered\, validated\, secured\
 , deployed and maintained. The overall challenge is thus to automate these
  activities accounting for the statistical nature of ML-enabled software.
  \n\nTaking a software engineering perspective\, and starting from a conc
 rete case from the finance industry\, the seminar will focus on the robust
 ification of a ML model which is integrated in a larger software system th
 at takes as input domain objects (e.g. financial transaction\, malware\, n
 etwork traffic). One traditional way to robustify a ML model consists in g
 enerating adversarial inputs\, e.g. leading to a  misclassification\, and
  retraining the model. Indeed\, despite their impressive performance\, ML 
 models are sensitive to small perturbations in the input. The resulting ad
 versarial inputs raise multiple questions about the robustness of such sys
 tems\, especially in safety- and business-critical domains. However\, the 
 generation of feasible\, exploitable adversarial examples is challenging\,
  as they must satisfy the business logic constraints over the feature spac
 e. We analyse the limitations of current adversarial approaches and explor
 e new algorithms that combine multi-objective search with constraint-solvi
 ng techniques. While the attack part is the offensive weapon\, we also con
 sider the challenge to efficiently shield the systems against such threats
 \, and finally end the seminar by mentioning other research directions to 
 deploy robust ML-enabled systems.\n\nBio: \nYves Le Traon is professor in 
 software engineering at University of Luxembourg\, with a focus on softwar
 e testing\, software security\, data-intensive and machine-learning (ML) e
 nabled systems. He is currently head of the SnT SerVal group (SEcurity\, R
 easoning and VALidation) and vice director of the Interdisciplinary Centre
  for Security\, Reliability and Trust (SnT center) in Luxembourg. Most of 
 his initial research has focussed on software testing\, software security 
 and model-driven engineering with applications to business-critical softwa
 re in the domains of Smart Grid\, Fintech\, IoT/Industry 4.0. His researc
 h interests also cover several related areas such as real-world empirical 
 studies\, search-based software engineering\, code static analysis\, autom
 ated-program repair\, machine learning applied to industry and robustness 
 of machine-learning enabled systems.\n\nIn terms of education\, he receive
 d his engineering degree and his Ph.D. in Computer Science from the “Ins
 titut National Polytechnique”\, Grenoble\, France\, in 1997. From 1998 t
 o 2004\, he was associate professor at the University of Rennes\, France. 
 From 2004 to 2006\, he was research engineer at “France Télécom R&D”
  company. In 2006\, he became full professor at Telecom Bretagne in France
  where he pioneered software testing techniques to find security flaws. He
  joined university of Luxembourg in 2009.\n\nProf. Le Traon\, IEEE Fellow 
 member\, has been General Chair of major conferences in the domain\, such 
 as ICSME 2021\, IEEE ICST 2013\, IEEE ISSRE 2004\, and Program Chair of IE
 EE QRS 2016. He serves at the editorial boards of international journals i
 n his domains of expertise (STVR\, SoSym\, IEEE Transactions on Reliabilit
 y). He has served in more than 60 program and organizing committees of con
 ferences\, including the flagship ones in software engineering and softwar
 e testing (ICSE\, FSE\, ICST\, ISSTA). He has co authored (CORE ranking): 
 80+ Rank A/A* conference papers\, 30+ Rank A/A* journal papers\, and 150+ 
 peer-reviewed conference papers\, 40+ journal papers and 12+ book chapters
  and magazines.\n 
LOCATION:ELA 2 https://plan.epfl.ch/?room==ELA%202 https://epfl.zoom.us/j/
 67632017825
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR