Artificial Intelligence workshops

The training school develops three foundations for AI in scientific computing. Daniel Bartl introduces statistical learning through high-dimensional probability, explaining how the geometry and complexity of function classes govern generalization. Stanislav Budzinskiy covers floating-point errors, deterministic and stochastic rounding, matrix multiplication, mixed precision and conditioning in neural-network training and inference. Maximilian Herde introduces neural operators and foundation models for partial differential equations, studying pretraining, scaling and transfer to new physical problems, with practical Python exercises. The school meets in person at RICAM, Science Park 2, fourth floor, Altenberger Straße 69, Linz, on 5–9 October. Monday through Thursday lectures start at 09:00 and run to 12:30, with discussion and exercise sessions generally scheduled 13:30–17:00; Wednesday has an excursion. Friday’s lectures run 09:00–11:30. Times are Europe/Vienna, CEST (UTC+02:00). The registration form still lists the Training School and requires an institutional email address. Registration is binding; participants should review the organizer’s cancellation and non-attendance policy. The event begins shortly after this listing’s publication.

This workshop connects mathematical and theoretical neuroscience approaches to the dynamics of densely connected recurrent spiking networks. Mean-field methods link individual-neuron dynamics with population behaviour, while low-dimensional collective dynamics and low-rank connectivity provide complementary descriptions informed by experiments and simulations. Researchers with mathematical and physics backgrounds alternate presentations to compare these perspectives. Public research lectures run from Monday afternoon through Wednesday, followed by doctoral tutorials on Thursday and Friday at the Bernoulli Center, EPFL, Lausanne. Registration begins at 13:30 on 2 November, with the introduction at 13:45 and talks from 14:00. Tuesday and Wednesday lectures and Thursday and Friday tutorials start at 09:00. Times are Europe/Zurich, CET (UTC+01:00). Seats are limited; researchers from outside EPFL pay CHF90 and commit to attending the lecture sessions. Tutorials are restricted to PhD students. Separate registration forms for doctoral students and other researchers are linked on the official workshop page. Preparation assumes mathematics or physics at master’s level and familiarity with computational neuroscience.

This RICAM workshop examines the numerical foundations of reliable deep-learning computation. Low-precision and mixed-precision arithmetic, quantization and parallel GPU computation enable larger models and deployment on constrained devices, while introducing questions about error, stability and performance. Researchers from theoretical and applied communities will compare methods for stable training and efficient inference and identify directions for further work. The programme brings together expertise in numerical analysis, machine learning and hardware-aware computation. The in-person workshop runs on 2–6 November at RICAM, Science Park 2, fourth floor, Altenberger Straße 69, Linz. Monday registration and welcome run 14:00–14:40, followed by talks until 17:00. Tuesday through Friday sessions begin at 09:00; Friday’s programme ends at 11:40. Times are Europe/Vienna, CET (UTC+01:00). Register through the special-semester form, select Workshop 2 and Regular Participant, and use an institutional email address to verify affiliation. Registration is binding; the organizer asks for cancellation at least one week before the workshop and states that unreported non-attendance can exclude participants from future events.