The EPFL Photonics Chapter is happy to invite you for a summer hike in the beautiful region of Leysin!
\r\nThe hike is planned to take place on Sunday June 14th 2026.
\r\nWe hope to give you a chance to meet and chat with your Photonics colleagues working across many different departments.
\r\nFor organizational purposes, please register for the event using this form before June 11h, 2026.
\r\n
\r\nThe travel costs starting from EPFL will be fully covered by the chapter, for those who have registered.
\r\nNote that if you register but eventually do not come*, we might ask you to reimburse the train ticket.
\r\n*Unless you cancel before June 11h at 23:59, by sending us an email.
\r\n
\r\nThe planned route will take us around the peak of Le Riondaz, offering beautiful views of the surrounding mountains.
\r\nThe hike is of moderate difficulty, with a few steeper sections along the way. However, the goal is for everyone to enjoy the day together, so we will keep a comfortable pace and make sure no one is left behind.
\r\nFurther details about the itinerary will be announced before the hike.
\r\n
\r\nFor any questions, please contact us at epc-committee@groupes.epfl.ch.
\r\nWe thank the Doctoral Program in Photonics (EPDO), the Society of Photo-Optical Instrumentation Engineers (SPIE.), Optica and SwissPhotonics for their support.
\r\n
The 2027 edition of the CROSS program calls upon researchers at EPFL and the University of Lausanne to submit proposals for joint projects that bring the natural sciences and engineering together with the social sciences and humanities.
\r\n
\r\nThrough this annual call for projects, CROSS provides competitive grants to support new seed research endeavors that have the potential to grow into full–scale interdisciplinary research projects.
\r\n
\r\nThe 2024 CROSS call is open to all topics.
\r\n
\r\n—————————————————————————————
\r\n
\r\nPROPOSALS:
\r\n
\r\nProposals must come from joint UNIL-EPFL teams, which bring together specialists in the human and social sciences on the one hand, with specialists from life sciences, natural sciences or engineering on the other. Up to six projects will be selected, with a maximum of CHF 60,000 awarded per project.\r\n
In combinatorics and ergodic theory, the decomposition into structured and random objects has been successfully exploited to prove density results and refinements. In both fields, nilsystems are the model for structured configurations. Recently, Kra, Moreira, Richter, and Roberson used dynamics in nilsystems to provide a density version of Hindman's theorem. In this talk, we delve into further sumset consequences for these two extreme cases and provide some examples of applications.
","image_description":"","creation_date":"2026-05-26T10:23:46","last_modification_date":"2026-05-26T10:25:57","link_label":"","link_url":"","canceled":"False","cancel_reason":"","place_and_room":"CM 1 517","url_place_and_room":"https://plan.epfl.ch/?room==CM%201%20517","url_online_room":"","spoken_languages":["https://memento.epfl.ch/api/v1/spoken_languages/2/?format=json"],"speaker":"Tristán Rádić (Northwestern University)","organizer":"","contact":"Felipe Hernandez Castro","is_internal":"False","theme":"","vulgarization":{"id":3,"fr_label":"Expert","en_label":"Expert"},"registration":{"id":3,"fr_label":"Entrée libre","en_label":"Free"},"keywords":"","file":null,"icalendar_url":"https://memento.epfl.ch/event/export/120849/","category":{"id":1,"code":"CONF","fr_label":"Conférences - Séminaires","en_label":"Conferences - Seminars","activated":true},"academic_calendar_category":null,"domains":[],"mementos":["https://memento.epfl.ch/api/v1/mementos/7/?format=json","https://memento.epfl.ch/api/v1/mementos/427/?format=json"]},{"id":72033,"title":"AI Center Seminar - AI Fundamentals series - Emanuel Tewolde - \"Benchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmas\"","slug":"ai-center-seminar-ai-fundamentals-series-emanuel-t","event_url":"https://memento.epfl.ch/event/ai-center-seminar-ai-fundamentals-series-emanuel-t","visual_url":"https://memento.epfl.ch/image/33328/200x112.jpg","visual_large_url":"https://memento.epfl.ch/image/33328/720x405.jpg","visual_maxsize_url":"https://memento.epfl.ch/image/33328/max-size.jpg","lang":"en","start_date":"2026-06-15","end_date":"2026-06-15","start_time":"14:30:00","end_time":"15:30:00","description":"The talk is jointly organized by the EPFL AI Center and the DLAB as part of the AI fundamentals seminar series.
\r\n
\r\nHosting professor: Prof. Robert West (DLAB)
\r\n
\r\nTitle
\r\nBenchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmas
\r\n
\r\nAbstract
\r\nIt is increasingly important that LLM agents interact effectively and safely with other goal-pursuing agents, yet, recent works report the opposite trend: LLMs with stronger reasoning capabilities behave _less_ cooperatively in mixed-motive games such as the prisoner's dilemma and public goods settings. Indeed, our experiments show that recent models---with or without reasoning enabled---consistently defect in single-shot social dilemmas.
\r\n
\r\nTo tackle this safety concern, we present the first comparative study of game-theoretic mechanisms designed to enable cooperative outcomes between rational agents _in equilibrium_. Across four social dilemmas testing distinct components of robust cooperation, we evaluate four families of mechanisms: (1) repeating the game for many rounds, (2) reputation systems, (3) third-party mediators to delegate decision making to, and (4) contract agreements for outcome-conditional payments between players. Among our findings, we establish that contracting and mediation are most effective in achieving cooperative outcomes between capable LLM models, and that repetition-induced cooperation deteriorates drastically when co-players vary. Moreover, we demonstrate that the mechanisms become _more effective_ under evolutionary pressures to maximize individual payoffs.
\r\n
\r\nBio
\r\n
\r\nEmanuel Tewolde is a fourth-year Computer Science PhD student at Carnegie Mellon University advised by Vincent Conitzer. His research focuses on algorithmic game theory, reinforcement learning, and LLM agents, with an emphasis on the safety, coordination, cooperation, and alignment of AI systems. Ultimately, he strives to understand how to enable artificial intelligence and humans to effectively achieve better social outcomes in strategic interactions.
\r\n
\r\nEmanuel’s work is supported by the Cooperative AI PhD Fellowship, and has previously received the AAAI 2025 Best Poster Award. He has previously worked on AI research agents with Meta FAIR, and researched ML methods for renewable energy systems with the Fraunhofer-Gesellschaft. Prior to CMU, he completed Master’s and Bachelor’s degrees in Mathematics at Imperial College London and the Technical University of Darmstadt respectively.
The development of new materials from renewable resources represents a win-win for humanity, allowing us to create useful products while also addressing critical ecological and environmental challenges. Proteins are particularly promising building blocks for this purpose. From a material science perspective, they offer immense potential for developing novel structures, composites, and other advanced materials. In this talk, I will present our research on the use of proteins and marine biomass to create a wide range of new materials. I'll discuss our bottom-up technologies and demonstrate how we have applied them to develop materials for various applications, including the controlled green syntheses of various nanoparticles and metallic structures, Light-emitting devices, Photothermal materials, Smart Wound dressings, Antibacterial coatings, and more. I will also highlight the significant role that Jellyfish have played as a renewable resource in many of these applications and the unique collaboration of our group with the Israel Electric Corporation, aiming to collect Jellyfish which is clogging the cooling turbines of the electrical power stations.
\r\n
\r\nBio: Prof. Shachar Richter is a faculty member in the Department of Materials Science and Engineering at Tel Aviv University, Israel, and the head of the department. He earned his PhD in Materials Science and Chemical Physics from the Weizmann Institute of Science, Israel. After graduation, he completed a postdoctoral fellowship and was later appointed an independent staff member (MTS) at Bell Laboratories and Agere Systems in New Jersey, USA.
\r\nIn 2001, he joined the School of Chemistry at Tel Aviv University, and in 2013, he moved to the newly established Department of Materials Science and Engineering. He was a visiting professor at Monash University in Australia in 2017 and a distinguished visiting scientist at the Commonwealth Scientific and Industrial Research Organization (CSIRO). His diverse research area focuses in developing innovative materials from renewable resources, such as jellyfish-based composites for medical and cosmetic uses, as well as working in molecular and bioelectronics, bio-nanocomposites, and new patterning technologies.
\r\nProfessor Richter has held several public roles, including serving as department head, chair of the \"Advanced Materials\" Cathedra, leading the MSc program in Materials Science and Engineering at Tel Aviv University, acting as President of the Israel Vacuum Society (IVS), and chairing the chemistry committee of the Standards Institute of Israel. He also serves on the editorial boards of various journals. Some of his patents have been licensed and are in the process of being commercialized, and he has co-founded two startups.
Supercurrent diode effects have recently attracted much attention for their potential applications in superconducting logic circuits, rectification applications and also in the study of fundamental symmetries of the underlying superconducting order parameter. In this talk, I’ll introduce the idea of finite momentum Cooper pairing (FMCP) in systems with strong helical spin-momentum locking and how this can naturally give rise to non-reciprocity in superconducting critical currents namely the Josephson diode effect (JDE). I’ll describe our work establishing the presence of a large JDE due to FMCP in lateral Josephson junctions formed from centrosymmetric van der Waals type-II Dirac semimetals 1T-NiTe2 and 1T-PtTe2. A theoretical analysis based on a Ginzburg-Landau type model suggests that the FMCP introduces a phase difference between the first (Cooper pair) and second (Cooper quartet) harmonic supercurrents, which manifests as the JDE. Measurements of the JDE as a function of the in-plane magnetic field and half-flux quantum oscillations in the JDE are in good agreement with the predictions, indicating the presence of large second harmonic supercurrents in the junction. Surface modifications are shown to strongly enhance the supercurrents in the junction, nearly by a factor of two, indicating the strong contribution of the topological surface states to the supercurrent transport. Overall, 1T systems provide an interesting platform for exploring the interplay of FMCP, JDE and topological superconductivity.
\r\n
\r\nBio: Dr. Pranava Keerthi Sivakumar completed his Bachelors and Masters in Chemistry with a minor in Physics from the Indian Institute of Science, Bengaluru, India in 2017. He then moved to Germany where he completed his Ph.D. at the Martin Luther University in Halle, graduating summa cum laude. During his Ph.D., he worked on studying the electrical transport properties of materials with non-collinear spin textures and unconventional superconductivity due to strong spin-orbit coupling. Most notably during his Ph.D., he uncovered the role of finite momentum pairing induced Josephson diode effect in topological materials with helical spin-momentum. He is currently a postdoctoral researcher at the NISE department in the Max Planck Institute of Microstructure Physics, focusing on the interplay of magnetization dynamics and supercurrent transport in superconductor-ferromagnet heterostructures.
The regulatory genome encodes a complex language that determines when and where genes are expressed, yet reading and writing this language remains a central challenge in genomics.
\r\n
\r\nTwo complementary approaches — CRISPR-CLEAR, a dense base-editing method that pinpoints individual nucleotides controlling gene activity, and DNA-Diffusion, a generative AI framework that designs entirely new cell-type-specific regulatory elements — offer powerful new ways to decode and engineer this language. Together, they have yielded clinically relevant discoveries, including noncoding variants driving CAR-T resistance in leukemia, and point toward a future of agentic AI that allows researchers to interrogate the regulatory genome in plain language.