Plenary lectures
BeneluxzaalAn overview of the three plenary lectures which will take place during the course of both conference days.
Tuesday 20 January 10.25 hrs - 11:10 hrs
Beneluxzaal
Plenary lecture by Petra Rudolf, University of Groningen
Title: What Still Keeps Me Curious: Open Problems from Surfaces to Space
After many years exploring fascinating topics of surface science and nanoscience, I would like to turn the spotlight not on what has been accomplished, but on what still puzzles me. In this talk, I will discuss a selection of open questions I believe are worthwhile to pursue further. These include the origins of triboelectric charging, the mechanisms of charge transfer upon contact of 2D materials, the elusive nature of photoluminescence in carbon quantum dots, the structure of water at interfaces, and the formation of water on dust grains in protoplanetary disks.
Tuesday 20 January 20:05 hrs - 20:45 hrs
Beneluxzaal
Plenary lecture by Iker Zuriguel, Universidad de Navarra
Title: Bottleneck flow: from grains to pedestrians and robots
Bottlenecks are everywhere in nature and daily life. We shake a saltcellar to get grains flowing onto our food, red blood cells squeeze through narrow vessels, and crowds funnel through stadium doors that can’t keep up with the surge of people pressing forward. It’s the same general process — the flow of many agents through narrow passages —appearing in many different circumstances and scenarios, most with important implications for industry, safety, and even health. In this talk, I will expose the intriguing, and sometimes counterintuitive, features observed in the flow of discrete systems through bottlenecks. By comparing the behaviors displayed by a series of systems including inert grains, suspensions, robots, sheep and even humans, we will disclose the main physical mechanisms that control bottleneck flow, thereby highlighting the value of a multidisciplinary approach to tackling complex scientific problems.
Wednesday 21 January 15:50 hrs - 16:35 hrs
Beneluxzaal
Plenary lecture by Lenka Zdeborová, École Polytechnique Fédérale de Lausanne
Title: Statistical Physics Perspective on Understanding Learning with Neural Networks
For over four decades, statistical physics has studied exactly solvable models of artificial neural networks. In this talk, we will explore how such models offer insights into deep learning and large language models. Specifically, we will examine a research strategy that trades distributional assumptions about data for precise control over learning behavior in high-dimensional settings. We will discuss several types of phase transitions that emerge in this limit, particularly as a function of data quantity. In particular, we will highlight how discontinuous phase transitions are linked to algorithmic hardness, impacting the behavior of gradient-based learning algorithms. Finally, we will present recent progress in learning from sequences and advances in understanding generalization in modern architectures, including the role of dot-product attention layers in transformers.