Prochain événements

EDBB grad students

Yang Zhao (Tang Lab) -- Metabolically Armored CAR-T Cells Counter Dysfunction and Promote Stemness for Solid Tumor Clearance

Olga Mitrofanova (Lutolf Lab) -- Bioengineered Human Intestine With in vivo-Like Complexity and Function

Session Chair: Remo Bättig (Barth Lab)

Mini-symposium format: (see web page)
Read more

Matthias Troyer, Technical Fellow and Corporate Vice President, Microsoft

While still in early development, quantum computers are already overturning our notions of computing by promising to solve certain problems that are intractable on any imaginable classical computer.  I will present simple guidelines identifying the promising quantum applications, which are in chemistry and materials science. First classically intractable academic simulations will need tens of thousands of qubits, but simulating molecules and materials with a classically unachievable precision will require  more than a million qubits. While developing a qubit architecture able to reach that scale we can make progress on these challenging problems with novel machine learning approaches.

Read more

Prof. Emanuela Del Gado, Georgetown University, USA

Elasticity, Rigidity and Rheology of Soft Particulate Gels
 

Bio: Emanuela Del Gado is Professor In the Department of Physics at Georgetown University in Washington DC, where she also currently serves as Director of the Institute for Soft Matter Synthesis and Metrology. She received her undergraduate degree (Laurea in Physics, cum laude) at the University of Naples "Federico II" in Italy, where she also obtained a PhD in Physics in 2001. She was a Marie Curie Fellow at the University of Montpellier in France and a post-doctoral researcher at ETH Zurich in Switzerland,  and held visiting positions at ESPCI Paris and MIT. Before joining Georgetown University as Associate Professor with tenure in 2014, Emanuela was a Swiss National Science Foundation (SNSF) Assistant Professor in the Department of Civil, Environmental and Geomatic Engineering at ETH Zurich. In 2016 and 2018 she was awarded a Chair Joliot and a Paris Science Chair at ESPCI Paris. In 2017 she became Georgetown University Provost’s Distinguished Associate Professor and was MIT - CEE C.C. MEI Distinguished Speaker. She was elected Fellow of the Royal Society of Chemistry in 2018 and Fellow of the American Physical Society in 2020. 
Read more

  Prof. Chris Wiggins

Titre : Data science @ the new york times

Résumé :
Le groupe Data Science du New York Times développe et déploie des solutions d'apprentissage automatique pour les problèmes de la salle de presse et des entreprises.
Recadrer les questions du monde réel en tant que tâches d'apprentissage automatique nécessite non seulement d'adapter et d'étendre les modèles et les algorithmes à des cas nouveaux ou spéciaux, mais aussi de connaître la bonne méthode pour le bon défi.
Je vais d'abord décrire comment non supervisé, supervisé et méthodes d'apprentissage par renforcement sont de plus en plus utilisées dans les applications humaines pour la description la prédiction et la prescription, respectivement.
Je me concentrerai ensuite sur les cas " prescriptifs ", en montrant comment les méthodes issues de la littérature sur l'apprentissage par renforcement et l'inférence causale peuvent avoir un impact direct dans les domaines suivants : l'ingénierie, les affaires eta prise de décision en général.

Biographie :
Chris Wiggins est professeur agrégé de mathématiques appliquées à l'Université de Columbia et scientifique en chef des données au New York Times. À Columbia, il est membre fondateur du comité exécutif du Data Science Institute, et du département de physique et de mathématiques appliquées ainsi que du département de biologie des systèmes, et est membre affilié de la faculté de statistique. Il est cofondateur et coorganisateur de hackNY (http://hackNY.org), une association à but non lucratif qui organise depuis 2010 des hackathons étudiants semestriels et le hackNY Fellows Program, un stage d'été structuré dans des start-ups de New York. Avant de rejoindre la faculté de Columbia, il a été instructeur Courant à NYU (1998-2001) et a obtenu son doctorat en physique théorique à l'université de Princeton (1993-1998). Il est membre de l'American Physical Society et a reçu le prix Avanessians Diversity Award de Columbia. Son livre à paraître "Data Science in Context : Foundations, Challenges, Opportunities", avec Alfred Spector, Peter Norvig et Jeannette M. Wing, sera publié par Cambridge University Press en 2022 et est disponible sous forme de projet en ligne sur https://datascienceincontext.com/. Son livre à paraître "How Data Happened : A History from the Age of Reason to the Age of Algorithms", avec Matthew L. Jones, sera publié par Norton Press en 2023.

Livres :
"Data Science in Context : Foundations, Challenges, Opportunities" (Cambridge Press).
https://www.amazon.com/Data-Science-Context-Foundations-Opportunities/dp/1009272209/ et "How Data Happened : A History from the Age of Reason to the Age of Algorithms" (Norton Press ; 21 mars 2023) https://www.amazon.com/How-Data-Happened-History-Algorithms/dp/1324006730/


Le Centre pour les systèmes intelligents de l'EPFL (CIS) est une collaboration entre IC, ENAC, SB; SV et STI qui réunit des chercheurs travaillant sur différents aspects des systèmes intelligents. En juin 2020, le CIS a lancé ses Colloques CIS avec des orateurs invités de renom.
Plus d'info
Read more

James Hackett, Ph.D., Group Leader, EMBL Rome, Monterotondo (IT)

WEEKLY EPFL BIOE TALKS SERIES

Abstract:
Control of chromatin structure and accessibility is a central mechanism for the regulation and propagation of gene expression programs. Histone modifications contribute to this mechanism, but their precise role in mediating quantitative transcriptional responses and cellular memory is incompletely understood. Indeed, it remains debated whether chromatin modification states directly trigger transcription or whether they simply reflect a consequence of expression across cellular and genomic environments. Understanding this is important to establish the function of dynamic epigenomic landscapes associated with cell identity and fate transitions. Here, I will first present our attempts to deconvolve chromatin regulation at the level of global epigenome (re)programming events during early mammalian development. I will then discuss how precision technologies can capture the context-dependent and quantitative function of chromatin modifications on transcriptional regulation and memory.

Bio:
PhD, 2010, University of Edinburgh, UK.
Postdoctoral research at the Gurdon Institute, University of Cambridge, UK.
Group leader at EMBL since 2016.
Joint appointment with the Genome Biology Unit.


Zoom link (with one-time registration for the whole series) for attending remotely: https://go.epfl.ch/EPFLBioETalks


Instructions for 1st-year Ph.D. students who are under EDBB’s mandatory seminar attendance rule:
IF you are not attending in-person in the room, please make sure to

1. send D. Reinhard a note before noon on seminar day, informing that you plan to attend the talk online, and
2. be signed in on Zoom with a recognizable user name (not a pseudonym making it difficult or impossible to be identified).

Students attending the seminar in-person should collect a confirmation signature after the talk - please print your own signature sheet beforehand (71 kB pdf available for download here).
 
Read more

Prof. José Bico, PMMH-ESPCI

Abstract: Drawing a flat map of the Earth is fundamentally challenging as continents unavoidably end up distorted. Reciprocally, complex natural shapes such as the delicate shape of Orchidea petals emanate from differential growth. From an engineering point of view, similar shape changes can be obtained when flat patches embedded with a network of channels are inflated. We will discuss different strategies involving stretchable elastomers or, conversely, stiff fabrics. Can we program the resulting 3D shapes? How robust are such inflated structures?
   

Biography: José Bico is associate professor at ESPCI where he teaches Fluids and Solids Mechanics. He leads with his colleagues Benoît Roman and Étienne Reyssat the MecaWet group at the PMMH laboratory. "Table top” experimentalist, his research activities range from interfacial hydrodynamics to slender structures Mechanics. Very active in science diffusion, he co-authored with his colleagues Étienne, Benoît and Étienne Guyon, “Hidden wonders”, MIT Press 2021.
 
Read more

Henrik Strahl, Centre for Bacterial Cell Biology, Newcastle University, UK

Antibiotics that target different steps of bacterial cell wall synthesis such as beta-lactams, vancomycin, and fosfomycin are generally assumed to induce cell lysis as their core antibacterial mode of action. This bacteriolytic process is catalysed by cell’s own wall-degrading enzymes (autolysins) that degrade peptidoglycan in an uncontrollable manner upon inhibition of cell wall synthesis. This runaway degradation leads to weakening of the cell wall sacculus until the cells are no longer able to withstand turgor and undergo lysis. However, cell lysis is insufficient to explain the rapid killing observed in Gram-positive bacteria with a thick cell wall. In fact, Gram-positive bacteria frequently lose viability at a much faster rate than they lyse. To study the cellular mechanisms behind the rapid, lysis-independent killing in Gram-positives, we analysed the mode of action of cell wall antibiotics at a single-cell level using predominantly fluorescence microscopic techniques. Our experiments carried out with Bacillus subtilis and Staphylococcus aureus revealed that inhibition of cell wall synthesis, surprisingly, triggers depolarisation of the cytoplasmic membrane that precedes and is independent of the lysis process. Using various fluorescence reporters and cellular assays, we found that the membrane depolarisation induced by cell wall-targeting antibiotics leads to energy starvation, extensive disturbances of cellular spatial organisation, and production of reactive oxygen species (ROS) that is accompanied by protein, lipid and DNA damage. These findings suggest that, rather than being a relatively simple process linked to osmotic lysis, the bactericidal activity of cell wall-targeting antibiotics is a complex cellular phenomenon that integrates autolysin-catalysed bacteriolysis with severe cellular disturbances and damages triggered by membrane depolarisation.

Read more

Beth L. Pruitt, Biological Engineering, Mechanical Engineering, and Biomolecular Science and Engineering, University of California Santa Barbara (US)

Abstract:
Hypertrophic Cardiomyopathy (HCM) is characterized by thickening of the left ventricular wall and hypercontractility and has been linked to mutations in the sarcomere motor protein β-myosin (MYH7). Using single cell mechanobiology studies, we examined how the effects of single point mutations propagate to change the contractile dynamics and cellular morphology (sarcomere spacing, spread area, myofibril alignment) of human induced pluripotent stem cell derived cardiomyocytes (hiPSC-CMs). We micropattern islands of adhesive protein to constraining the spreading and alignment of hiPSC-CM on hydrogel substrates containing fluorescent microbeads as fiducial markers for traction force microscopy (TFM). We deployed substrate stiffnesses ranging from physiological (10 kPa) to heavily diseased/fibrotic (100 kPa) to test the role of increased “afterload” in functional phenotypes. We use image and video analysis to assess the contractile dynamics of the hiPSC-CM in terms of force, power, and velocities of relaxation and contraction. For example, we assessed multiple MYH7 mutations edited into the WTC line along with isogenic controls. Some lines carried an endogenously labeled alpha-actinin GFP reporter of sarcomere structure to enable visualization of sarcomere structure and dynamics. We assessed the magnitude and dynamics of contractile force output from TFM video analysis and observed increased the contractile force when compared to the control hiPSC-CMs. We also measured significantly different dynamics in the relaxation or contraction velocities compared to control hiPSC-CMs. Interestingly, not all HCM mutant lines presented a significant increase in cell spread area, a proxy for hypertrophy, and this correlated with culture conditions, such as the size of the protein pattern constraining the cells or stiffness of the substrate. Taken together, these results suggest a role for MYH7 mutations driving remodeling of structure and function at a cell-intrinsic level via changes in mechanosignaling.

Bio:
Dr. Beth Pruitt graduated from the Massachusetts Institute of Technology (MIT) with an S.B. in mechanical engineering. She was supported by a Navy ROTC fellowship at MIT where she learned sailing, leadership, and perseverance. She earned an M.S. in Manufacturing Systems Engineering from Stanford University before serving as an officer in the U.S. Navy. Her first tour was at the engineering headquarters of the Navy nuclear program providing engineering review and oversight to refueling operations. Her second tour was as at the U.S. Naval Academy as an instructor teaching Systems Engineering during the academic year and offshore sailing in the summer. She earned her Ph.D. in Mechanical Engineering at Stanford University where she specialized in MEMS and small-scale metrologies for electrical contacts and was supported by a Hertz Foundation Fellowship. She was a postdoctoral researcher at the Swiss Federal Institute of Technology Lausanne (EPFL) where she worked on polymer MEMS. Dr. Pruitt founded and led the Microsystems Lab at Stanford for 15 years, with research focused on small-scale metrologies for interdisciplinary micromechanics problems in mechanobiology, biomechanics and sensing. She was a visiting professor in Prof. Viola Vogel's Lab for Applied Mechanobiology in the Department of Health Sciences and Technology at ETH, Zurich in 2012. Dr. Pruitt moved to UC Santa Barbara in 2018 to help launch a biological engineering degree program and department. She has been Director of the Center for Bioengineering since 2019. She is an elected Fellow of BMES, AIMBE, and ASME and Senior Member of IEEE. She has been recognized by the NSF CAREER Award, DARPA Young Faculty Award, Denice Denton Leadership Award.
Read more

Dr. Daniel Sage

Our in-house expert Dr. Daniel Sage will give an Introduction Workshop on ImageJ/Fiji during our next imaging lunch. Open to all EPFL PhD students and Postdocs!Dr. Sage will go over the basics of ImageJ/Fiji, show us how to get to grips with the tool, present some useful plugins, give a short introduction on how to write macros, and present an example of how to build a typical image-analysis pipeline with the program.

Registration required

About the Imaging Lunches: Once per month, the EPFL Center for Imaging organises an event dedicated to all PhD students and postdocs working with/in imaging. Discuss the latest advances in imaging. Connect with imaging peers. Learn about popular imaging tools!

More information on the EPFL Center for Imaging: imaging.epfl.ch

Read more

Gerd Kortemeyer, ETHZ

The pandemic prompted significant shifts in the mechanisms of university teaching. 

Approaching the tail end of the pandemic, students have increasingly been given greater flexibility on how (and if) to attend lectures; whether on campus (in person), synchronously online, or asynchronously online (simply by watching recordings).

Faculty have thus rethought how they deliver lectures and blended, hybrid, or flipped scenarios are increasingly being considered.

The question now is where are we going with this, and what difference will it make? 

In this session of lunch&LEARN, Dr Gerd Kortemeyer, Director of the Educational Development and Technology (LET) department at ETH Zurich and Associate Professor Emeritus at Michigan State University, will present studies conducted with 285 students in an introductory physics course at MSU, and with 17,641 students and 639 faculty members at ETH Zurich. 

Kortemeyer will go over his findings, which indicate that there is no significant difference between in-person and online attendance, particularly when it comes to exam performance. 

Rather, the findings seem to indicate that it is our understanding of teaching and learning in higher education as we know it that may be undergoing subtle changes.

Please note that this event is internal and geared towards the EPFL teaching community. 
Registration is required to attend (click here to register)

Read more

La conférence sur les systèmes de traitement de l'information neuronale (NeurIPS) est la principale conférence sur l'apprentissage automatique.

Bien que la conférence soit très sélective, les chercheurs de l'EPFL ont 41 articles acceptés pour être présentés à la conférence en 2022.

Nous avons décidé, suite à la conférence aux Etats-Unis, d'organiser un événement local de taille moyenne sur le campus de l'EPFL, et d'inviter toute personne ayant une contribution acceptée à NeurIPS à postuler pour l'un de nos créneaux de présentation et/ou de poster. Les chercheurs de toutes les institutions (pas seulement de l'EPFL) sont invités à postuler et/ou à participer à l'événement. Veuillez noter que les frais de voyage ne peuvent pas être remboursés.

La conférence donnera l'occasion à tous les étudiants et chercheurs de l'EPFL dont les articles ont été acceptés à NeurIPS 2022 de présenter leurs travaux, et à tous les étudiants et chercheurs intéressés par la recherche en apprentissage automatique de se connecter et de discuter de la science pendant cet événement d'une journée.

Comité d'organisation :
Prof. Florent Krzakala, EPFL (Information, Learning & Physics Lab.) 
Prof. Martin Jaggi, EPFL (Machine Learning & Optimization Lab.)  
Prof. Pascal Frossard,  EPFL (Signal Processing Lab.)
Dr Jan Kerschgens, EPFL (CIS)

Date : 14 décembre 2022
Lieu : CO 2

L'événement dispose d'un nombre limité de places et d'emplacements pour les posters.  Veuillez vous inscrire à l'avance en utilisant le lien sur notre site web.
Programme à venir.

Read more

  Prof. Edoardo Mazza Experimental Continuum Mechanics, ETH Zürich
Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf

Abstract: The mechanical properties of human skin are linked to its function at both tissue and cell length scale. In fact, the tissue has to provide sufficient compliance to enable body movements, but it also forms a mechanically stable barrier against external loads. At cell length scale, the mechanical properties of the extracellular matrix influence the behavior of dermal cells, e.g. during tissue repair and skin growth.
We combined ex-vivo multiaxial tensile experiments with in-vivo suction measurements and 3D tissue imaging in order to develop a multilayer poroelastic model of human skin. Each skin layer is represented as a biphasic material, with the solid part characterized by a fiber network and a compressible matrix, while interstitial fluid flow is driven by gradients of the chemical potential, which result from the boundary conditions imposed and the fixed charge distribution in the tissue. The corresponding mechanical response indicates an average stiffness akin to a modulus in the range of 100 kPa. However, testing on the macroscale does not allow characterizing the mechanical microenvironment of dermal cells, for which several orders of magnitude lower stiffness has been reported.
We rationalized the discrepancy between micro- and macroscale mechanics using a hybrid discrete-continuum model representative of the heterogeneous microstructure of the dermis. Fibers are modeled as nonlinear elastic connectors. Biphasic continuum elements provide a representation of interstitial fluid, proteoglycans and other non-collagenous ECM components. Model parameters were selected to provide a reasonable fit for experimental data at macro- and microscales.
The resulting multiscale model representation of skin allows to investigate the relationship between tissue microstructure and its fracture properties, and specifically to understand the deformation mechanisms contributing to its high defect tolerance. Moreover, simulation of skin stretch in-vivo provides quantitative information on the associated changes in cell-perceived stiffness and chemical potential of the interstitial fluid, and both are expected to influence the behavior of resident cells during skin homeostasis and repair.
Read more

Are you a student entrepreneur yourself? Are you curious to discover disruptive entrepreneurial projects and startups from EPFL students?

Come to get a taste of EPFL’s entrepreneurship & innovation programs, while discovering the latest projects and rising startups! Join us for the Changemakers and blaze award ceremony to discover the next generation of EPFL startups!

During the event, the EPFL students who have followed the programs will pitch their startup projects and startup to the audience.

The Changemakers and the blaze accelerator programs support EPFL students drive future change through their entrepreneurial projects. The pitching competition is the final event of a semester-long support program including workshops, coaching and networking events to take these startups to the next level. 

Join us at the SPOT for the Changemakers and blaze award ceremony on Thursday 15 December 2022.
Participation is free but registration is required.

Program:
17:15 Doors open
17:30 Start of the event and pitching
19:00 Awards followed by an aperitif

Changemakers is a 10-week learning journey for bachelor, master and PhD students to boost their innovation and entrepreneurial skills whilst turning their idea into a real-life project.
For any interest in joining the next cohort in 2023, contact Marius

blaze is an accelerator program designed to advance student startups towards a successful market launch, by transforming founders into leaders and projects into startups.
If you are interested to join the next blaze cohort in 2023, contact Maurice

Read more

There is an opportunity to leverage the intersection between AI, Robotics, and Lab Sciences to accelerate discoveries and findings. This requires timely interdisciplinary action from research and industry in the domains of AI and Machine Learning, Robotics and Automation, and Lab Sciences. This workshop intends to bring together experts in these fields to discuss cutting-edge research, trends, and future opportunities in this synergistic area.

We invite you to register to participate in this event.  For students (masters/PhD/Post-doc) we also invite you to register to present a poster and short flash talks.  There will be an award for the best presentation. 
 
To view the list of invited speakers and the provisional schedule visit here.
To register, please click here DEADLINE FOR REGISTRATION December 9th.

PROVISIONAL SCHEDULE:
08:45Arrival & Coffee
09:00Introduction & Welcome 
09:10 Session 1 Academia Keynote Speakers + Round Table

• Pascal Miéville – SwissCAT+
• Philippe Schwaller – Laboratory of Artificial Chemical Intelligence (EPFL)
• TBC
• Round Table Discussion Session 1 Speakers

10:40Coffee Break
11:00Session 2: Robotics for Lab Automation

• Soheil Gholami – Learning algorithms and systems laboratory (EPFL)
• Olga Fink – Intelligent Maintenance and Operations Systems (EPFL)
• Selman Sakar – MicroBioRobotic Systems Laboratory (EPFL)
• Round Table Discussion Session 2 Speakers

12:30Lunch & Posters
13:30Student Flash Talks 
14:00Future Vision for Lab Automation: Speakers TBC
14:30Session 3: Industry Perspectives

• Loïc Roch – Atinary 
• Patrick Courtney – Sila
• Enrico Eberhard – AICA 

15:15Coffee Break
15:30

• Oliver Peter – Idorsia
• Soft Robotics Lab (ETH Zürich)

16:00Industry Panel Discussion
16:30Wrap up & Next Steps 

• Pascale Van Landuyt – EPFL Vice Presidency for Innovation (VPI) 
• Michael Opieczonek – NTN innovation booster 
• Prizes for best talk & best student poster

16:45End of event
Read more

J-P. Kneib (EPFL), L. Valenziano (Italian GC in Zürich), F. Avino (EPFL), I. Furno, F. Romano (EPFL), E. Chesta (CERN), G. Becker (Beyond Gravity), L. Piguet (Clearspace), D. Campi (Sydereus Space Dynamics) , T. Montaruli (Univ. Geneve), E. David., B. Gresse (TAS - CH), C. Colombo (PoliMi) V. Striano (DAC), G. Sylos Labini (AIPAS), A. Fasoli (EPFL)

This workshop is free but registration is mandatory for organisation purposes. Register HERE.
 
Read more

Prof. Theodore (Ted) S Rappaport
Founding director NYU WIRELESS

Abstract
Future wireless cellular networks will utilize millimeter wave and sub-THz frequencies and deploy small-cell base stations to achieve data rates on the order of hundreds of Gigabits per second per user. The move to sub-THz frequencies will require attention to sustainability and reduction of power whenever possible to reduce the carbon footprint while maintaining adequate battery life for the massive number of resource constrained devices to be deployed. This article analyzes power consumption of future wireless networks using a new metric, the power waste factor (W), which shows promise for the study and development of “green G” - green technology for future wireless networks. Using W, power efficiency can be considered by quantifying the power wasted by all devices on a signal path in a cascade. We then show that the consumption efficiency factor (CEF), defined as the ratio of the maximum data rate achieved to the total power consumed, is a novel and powerful measure of power efficiency that shows less energy per bit is expended as the cell size shrinks and carrier frequency and channel bandwidth increase. Our findings offer a standard approach to calculating and comparing power consumption and energy efficiency in cascaded systems. Finally, we postulate this framework could be applied to other aspects of power efficiency metrics for data centers, base stations, algorithmic design, and software design.

Bio
Theodore (Ted) S. Rappaport (tsr@nyu.edu) is the David Lee/Ernst Weber Professor in Electrical and Computer Engineering at New York University (NYU), and is a professor in the NYU Courant Computer Science Dept. and the NYU School of Medicine. He founded the NYU WIRELESS research center and the wireless research centers at the University of Texas Austin (WNCG) and Virginia Tech (MPRG). His work has provided fundamental knowledge of wireless channels and system design for the first IEEE 802.11 Wi-Fi standard, the first U.S. digital TDMA and CDMA standards, the first public Wi-Fi hotspots, and proved the viability of millimeter wave and Terahertz frequencies for 5G, 6G, and beyond. He founded two businesses that were sold to publicly traded companies – TSR Technologies, Inc. and Wireless Valley Com­munications, Inc., and was an advisor to Straight Path Communications which sold 5G millimeter wave spectrum to Verizon. He is a licensed Professional Engineer and is a member of the Wireless Hall of Fame, the US National Academy of Engineering, a Fellow of the US National Academy of Inventors, and a life member of the American Radio Relay League. His amateur radio call sign is N9NB.
 
Read more

Prof. Thomas Pardoen, Institute of Mechanics, Materials and Civil Engineering, UCLouvain

Abstract: The race for ultra-high strength in metallic alloys has turned into an even more complex quest for high strength/high ductility balance over the last two decades. Depending on the list of requirements of the applications, a minimum ductility is always needed at least for formability. More recently, it has been recognized that a high strength / high ductility compromise does not necessarily translate into high fracture toughness, involving the remnant confusion between the concepts of toughness and fracture toughness. A vast number of structural applications calls for an optimum trade-off between fracture toughness, strength and ductility with a weight that depends on the type of loading conditions and constraints. This has brought back to the front scene the old question of the link between fracture toughness and the two other properties.

In the simplest paradigm, fracture resistance expressed in terms of critical J integral J_Ic or critical energy release rate G_Ic is a product of strength, true fracture strain under the appropriate triaxiality and a microstructure length scale. The last aspect is often neglected as well as other contributors, such as a possible change of failure mechanism at crack tip compared to tensile bars, the resistance to void nucleation, the strain hardening capacity and the thickness dependent dissipation by crack tip necking in plates and sheets. Recent results on DP steels with elongated martensite second phases, Cantor type high entropy alloy, Ti-12wt.% Mo alloy, friction stir modified Al alloys and stainless steel, will highlight several important messages on how to raise or to keep high fracture toughness in strong and ductile systems. 

Biography: Thomas Pardoen is full professor at the Ecole Polytechnique de Louvain and at the Institute of Mechanics, Materials and Civil Engineering of UCLouvain. He is the Senior Advisor to the President of UCLouvain for corporate relations. Outside UCLouvain, he is the Chair of the Scientific Council of the Belgian Nuclear Research Center SCK•CEN, vice chair of the Board of the Von Karman Institute (VKi) and of the Board of the Centre Terre et Pierre (CTP). He represents Belgium at the Euratom Science and Technical Committee and at the Global Nuclear Forum (NEA/OECD). After graduating as engineer (1994), receiving a master in philosophy (1996) and a PhD (1998) at UCLouvain, and being a postdoctoral researcher at Harvard University, he became faculty member in 2000. His research interests span the area of the nano-, micro- and macro- mechanics of materials and systems, with an emphasis on multiscale experimental investigations and modelling of deformation and fracture phenomena, as well as coupled functional-mechanical properties and irradiation effects, from both fundamental and applied perspectives. His research activity is articulated around the mechanics of (i) composites, hybrids, multimaterials, and adhesives, (ii) thin films, coatings and mems, (iii) high performance metallic alloys. He has supervised >50 Ph. D. students and > 25 post docs. He is a member of the editorial advisory board of J. Mech. Phys. Solids, Engng. Fract. Mech and Int. J. Damage Mech. He has published over 230 papers in peer reviewed international journals, with current h factor = 66 (Google), and 4 patents. He received the Grand Prix Alcan of the French academy of sciences in 2011 and a Francqui Chair from Université de Liège in 2015. He has been nominated Euromech Fellow in 2015.
Read more

Prof. Tanja Weil, Max Planck Institue, Germany


Read more

Prof. Ulugbek Kamilov, Washington University in St. Louis

Abstract
Computational imaging is a rapidly growing area that seeks to enhance the capabilities of imaging instruments by viewing imaging as an inverse problem. Plug-and-Play Priors (PnP) is one of the most popular frameworks for solving computational imaging problems through integration of physical and learned models. PnP leverages high-fidelity physical sensor models and powerful machine learning methods to provide state-of-the-art imaging algorithms. PnP models alternate between minimizing a data-fidelity term to promote data consistency and imposing a learned image prior in the form of an “image denoising” deep neural network. This talk presents a principled discussion of PnP and recent results on PnP under inexact physical and learned models. Inexact models arise naturally in computational imaging when using approximate physical models for efficiency or when test images are from a different distribution than images used for training. We present several successful applications of our theoretical and algorithmic insights in bio-microscopy, computerized tomography, and magnetic resonance imaging.

Biography
Ulugbek S. Kamilov is the Director of Computational Imaging Group and an Assistant Professor of Electrical & Systems Engineering and Computer Science & Engineering at Washington University in St. Louis. He obtained the BSc/MSc degree in Communication Systems and the PhD degree in Electrical Engineering from EPFL, Switzerland, in 2011 and 2015, respectively. From 2015 to 2017, he was a Research Scientist at Mitsubishi Electric Research Laboratories, Cambridge, MA, USA. He is a recipient of the NSF CAREER Award and the IEEE Signal Processing Society’s 2017 Best Paper Award. He was among 55 early-career researchers in the USA selected as a Fellow for the Scialog initiative on “Advancing Bioimaging” in 2021. His PhD thesis was selected as a finalist for the EPFL Doctorate Award in 2016. He has served as a Senior Member of the Editorial Board of IEEE Signal Processing Magazine and as an Associate Editor of IEEE Transactions on Computational Imaging. He has served on IEEE Signal Processing Society’s Computational Imaging Technical Committee and Bioimaging and Signal Processing Technical Committee. He was a plenary speaker at iTWIST 2018 and is a program co-chair for the International Biomedical and Astronomical Signal Processing Frontiers conference for 2023.

Register here!
 
Read more

Ingrid Le Duc

Being a good lecturer is an achievement, not an innate gift.  Join this workshop to identify your potential as a speaker and lecturer, to better plan in advance without wasting your time and in doing so keep your spontaneity and pleasure in teaching.

Read more

Prof. Jean-François Remacle

The Civil Engineering Institute, in collaboration with ENAC Open Science and the Open Science office, are happy to present:

The X-MESH method for capturing interfaces
by Prof. Jean-François Remacle, from UCLouvain  
 
In this presentation, we develop an innovative approach - X-MESH - to overcome a major difficulty associated with numerical simulation in engineering: we aim to provide a revolutionary way to track physical interfaces in finite element simulations. The idea is to use so-called extreme mesh deformations. This new approach should allow low computational cost simulations as well as high robustness and accuracy. X-MESH is designed to avoid the pitfalls of current ALE methods by allowing topological changes on fixed mesh.
The key idea of X-MESH is to allow elements to deform until they reach a zero measure. For example, a triangle can deform into an edge or even a point. This idea is rather extreme and completely revisits the interaction between the meshing community and the computational community, which for decades have been trying to interact through beautiful meshes.
In this talk, we will focus on both the mathematical issues related to the use of zero-measure elements and the X-MESH resolution scheme.Several applications will be targeted: the Stefan model of phase change, two-phase flows and contact between deformable solids.

The seminar will be followed by an open discussion on the benefits and obstacles of developing open science projects, and how to rethink the future of research in a more collaborative way. Food will be provided following the seminar, to encourage networking and get to know the speaker!

About the speaker
 
After his Engineering Degree at the University of Liege in Belgium in 1992, Jean-François Remacle obtained in 1997 a Ph.D. from the same University. He then spent two years at the Ecole Polytechnique de Montréal as a post-doctoral fellow of Prof. F. Trochu, followed by three years at Rensselaer Polytechnic Institute in the research team of Prof. M. Shephard (one year as research associate followed by two years as research assistant professor). It was during his stay at Rensselaer that Pr. Remacle started to work closely with Mark Shephard on mesh generation. Pr. Shephard’s seminal work on mesh generation is one of the most important contributions ever. It was also during that stay that Pr. Remacle started the development of Gmsh, the open source mesh generator. After these five years in Northern America, Jean-François Remacle joined the Université catholique de Louvain in 2002 as an assistant Professor. He then became Associate Professor in 2005 and Full Professor in 2012. In the following years of his return to Europe, Pr. Remacle dedicated a large part of his research to mesh generation.
 
Since 2002, Pr. Remacle and his colleague Pr. C. Geuzaine from the University of Liège have continued the development of Gmsh (www.gmsh.info). Gmsh was initially released as an open source in 2003 under the GNU General Public Licence (GPL). In 2009, a paper was published in the International Journal for Numerical Methods in Engineering (IJNME) that describes original features of Gmsh [GR09]. This paper is the most cited paper of IJNME in the last 3 years (https://onlinelibrary.wiley.com/journal/10970207). Gmsh was awarded a free software prize at the “trophées du libre” in 2009 . The size of Gmsh’s user community is now of over 8,000 regular users, including engineers of major European industries like Siemens, Dassault, EDF, Airbus or Snecma. Three Gmsh workshops have been organized, the last one in 2017.
 
In 2015, Pr. Remacle received an ERC Advanced Grant (www.hextreme.eu) with two major subjects: fast mesh generation and hexahedral mesh generation. Several breakthroughs have been achieved in HEXTREME, the three most significative ones being the developement of the fastest tetrahedral mesh generator, the use of Ginzburg Landau theory for generating quad meshes, and the development of an algorithm to build combinatorial hexahedral meshes whose boundary facets exactly match a given quadrangulation of the topological sphere.
 
In 2022,  Pr. Remacle received an ERC SYNERGY Grant together with Pr. Moës (Ecole Centrale de Nantes).
 
Since 2016, the papers co-authored by Jean-François Remacle received over 6600 citations (Source: Google Scholar). It is interesting to note that these citations are not only related to mesh generation but to a wide a spectrum of computational fields: computational fluid dynamics, ocean modeling, computational material science, fracture mechanics, biomechanics, scientific visualization, high performance computing.
 
In parallel to the fundamental developments of the mesh generation, Pr. Remacle has been actively involved in collaborative projects. The fruitful interactions that he had with engineers of major European consortia (Airbus, Siemens, Dassault) have convinced him of the great interest of the European industry for new  developments in mesh generation for computational mechanics.
 
 

Read more

Ruth Graham; Breakout Groups

The teaching day is an annual event organised jointly by CCE, CAPE, and LEARN. It is the general assembly of the CCE, and it provides the opportunity to the teaching body of EPFL to meet and discuss key-topics related to teaching at EPFL. 

 

Read more

Ingrid Le Duc

Pour concevoir son enseignement dans une démarche cohérente et bouclée d’ingénierie pédagogique et pour réfléchir à comment choisir et mettre en œuvre des méthodes et une organisation de son enseignement de manière à favoriser les acquis d’apprentissage des étudiants. With the use of concept maps you can structure a course that integrates learning outcomes, course content and students’ learning activities.

Read more