The talk is jointly organized by the EPFL AI Center and the EPFL Machine Learning for Biomedicine Lab (MLBIO) as part of the AI Fundamentals seminar series.
\r\n
\r\nTitle
\r\nAgents That Know What They Know: Interpretability and Introspection for Lifelong Learning
\r\n
\r\nAbstract
\r\nOne of the hallmarks of intelligence is the ability to learn from experience. In artificial intelligence, we frame this as lifelong learning: autonomous agents that continue to adapt after deployment as they encounter new tasks and environments. In this talk, I will argue that successful lifelong learning agents require a form of introspection: they must be able to reason about the scope and limits of their own knowledge. This self-knowledge allows an agent to answer three fundamental questions. What do I know? When can I reuse what I know? And what do I still need to learn? I will present recent work from my group showing how ideas from interpretability and explainability can surface the knowledge needed to answer these questions. I will first show how learned representations can be used to identify when inputs are novel, enabling out-of-distribution detection. Then, I will discuss how this can be built upon to give agents a way to distinguish between situations where prior knowledge can be safely reused and situations where new learning is needed. Together, this work shows that interpretability is not only a tool for transparency, but a mechanism for building agents that recognize, reuse, and acquire knowledge over time.
\r\n
\r\nBio
\r\nJoseph Campbell is an Assistant Professor in the Department of Computer Science at Purdue University, where he leads the Collaborative AI for Machines and People Lab. His research bridges machine learning and robotics, with a focus on explainable machine learning and lifelong learning. Before joining Purdue, he was a Postdoctoral Fellow in the Robotics Institute at Carnegie Mellon University, and he received his Ph.D., M.S., and B.S. degrees from Arizona State University. His research is supported by Amazon, the Toyota Research Institute, and the National Science Foundation.
The EPFL AI Center is delighted to invite you to join the exchange session about the Research Computing Platform (RCP).
\r\n
\r\nAre you building, training ML models? Are you currently facing challenges regarding IT infrastructure, and looking for reliable, collaborative, high-capacity, and affordable storage solutions to manage your research data? Are you seeking information about the computing costs?
\r\n
\r\nThe Research Computing Platform (RCP) offers campus-wide IT infrastructure geared for the research community. The offering includes a GPU infrastructure (> 400) aiming at making AI compute as accessible as possible throughout the campus.
\r\n
\r\nThis event will provide insights into the various services RCP offers to the EPFL research community.
\r\n
\r\nEvent Details:\r\n
Le stress et l’anxiété font aujourd’hui partie intégrante du quotidien, d’autant plus dans un cadre professionnel. Pression de performance, surcharge d’informations, incertitudes et distractions digitales, nous affectent toutes et tous. Lorsqu’ils deviennent chroniques, ces états impactent la santé mentale, le sommeil, la concentration, la motivation, les relations sociales et la qualité du travail.
\r\n
\r\nComprendre ces mécanismes, apprendre à les réguler, et surtout identifier ses propres déclencheurs et ressources devient alors essentiel pour évoluer dans un cadre professionnel et promouvoir un travail stable et une qualité de vie plus saine.
\r\n
\r\nCette formation vise à renforcer votre résilience, vous permettre d’identifier vos déclencheurs personnels, et d’activer des ressources internes et externes pour retrouver calme, clarté mentale et stabilité au quotidien.
European Framework Programs for Research and Innovation.
\r\n
\r\nIs your laboratory involved in one or several European projects, and is one of your responsibilities to ensure their effective administrative, financial, and human‑resources management? If so, it is important to take the time to become familiar with the structure of these projects and with the specific management rules that apply to them.
\r\n
\r\nThis training course will guide you through the contractual documents and their relevant content for project management. Administrative, financial, and HR management rules will be presented and discussed using practical case studies.
\r\nThe training will also help participants become familiar with the European portal and the SEFRI reporting platform.
\r\n
\r\nIt is recommended to attend this training as early as possible, ideally at the start of the project.
Thesis Director: Prof. S. C. Olhede,
\r\nMathematics doctoral program
\r\nThesis Nr. 11276
\r\n
\r\nTo take part in the public defense, please contact directly the speaker
Thesis Director: Prof. W. Gerstner,
\r\nNeuroscience doctoral program
\r\nThesis Nr. 11158
\r\n
\r\nTo take part in the public defense, please contact directly the speaker
Stein's Restriction conjecture concerns functions whose Fourier transform is supported on the unit sphere in R^n. Over the decades, progress on this problem has drawn on tools from combinatorics, real algebraic geometry, and other areas. We will survey the development of the conjecture and discuss its recent connections to projection theory.
\r\n
\r\n
Join us for the 9th edition of the Sustainable Space Hub (SSH) Coffee, an opportunity to connect with the community and explore the latest developments in space sustainability.
\r\n
\r\nStephan Hellmich will present the new SSA (Space Situational Awareness) activities: the NASO (Extracting Space Object Observations from large Astronomical Data Archives) and TESSA (Technologies to Enhance Space Situational Awareness) projects.
\r\n
\r\nIn collaboration with the University of Bern, the TESSA project continues to advance capabilities in satellite tracking and space object monitoring. Work in 2026 pursues two parallel research tracks: refining 6D pose estimation techniques for the precise determination of object position and orientation in space, and developing adaptive optics solutions for the Zimmerwald Observatory, including the associated algorithms for correcting atmospheric distortion. Together, these efforts contribute to the broader challenge of space situational awareness and safe orbital operations.
\r\n
\r\nDuring the event, the space semester projects for the fall semester will also be announced.
\r\n
\r\nProvisional schedule
\r\n12:30 – Welcome, coffee, cookies & networking
\r\n13:00 – Stephan Hellmich will present the new SSA activities: the NASO and TESSA projects.
\r\n13:10 – Presentation of the space semester projects for the fall semester.
\r\n13:45 – Q&A and open discussion
\r\n
\r\nWe look forward to welcoming you for an engaging discussion and informal networking around the latest advances in sustainable space activities.
\r\n
\r\nMore information
\r\nSemester projects: https://espace.epfl.ch/education/student-projects/
\r\nPresearch at EPFL Space Center: https://espace.epfl.ch/research/