Physics of Intelligence and Machine Learning 2024

Venue and Dates

Universität Heidelberg
Im Neuenheimer Feld 225a (3rd Floor)  EINC 03.128
69120 Heidelberg, Germany
(Open in Google Maps)

From 14th to 18th October, 2024

Descriptions

In 2023, we organized two small-sized workshops in Japan:

1. Topology and Machine Learning: Workshop in Shonan 2023

2. A Focused Week of Physics for Machine Learning in Kyoto 2023

These workshops were extremely successful with active discussions in a friendly atmosphere.  There were lectures from the Japan team and the Heidelberg (+Munich) team and talks followed with ample time for discussions.  The aim was to update us with the state-of-the-art developments in machine learning theory/applications and pursue the possibility of utilizing our knowledge about theoretical physics such as the quantum field theory, the classification by topological invariants, etc. for improving machine learning. 

To continue the active discussions and enhance the perspective, we plan to have a larger-sized workshop keeping the informal and friendly atmosphere, this time in Heidelberg! 

The workshop will have lectures and talks followed by a whole day reserved for discussions.

We cannot overemphasize the importance of physics in the development of machine-learning technologies.  The Nobel-Prize press-release states: They trained artificial neural networks using physics.  Indeed, physics is such a universal language and we are hoping to get wiser with the physics of intelligence!

Participation is by invitation only.
If you are interested, please contact the organizers (Kenji Fukushima / Jan M. Pawlowski).

Invited Lecturers

• Tristan Bereau (Heidelberg U.) -- Generative ML for molecular simulations

• Ulrich Köthe (Heidelberg U.) -- Free-form flows and their applications

• Fred Hamprecht (Heidelberg U.) -- ML for density functional theory

• Koji Hashimoto (Kyoto U.) -- Symmetries in Neural Networks

• Theo Heimel (U. Louvain) -- MadNIS: future ML-event generator

• Tilman Plehn (Heidelberg U.) -- Generative models

• Robert Scheichl (Heidelberg U.) -- Bayesian inference, inverse problems & multiscale analysis

• Masato Taki (Rikkyo U.) -- Basics and applications of Transformer

• Akinori Tanaka (RIKEN) -- Towards “Physics” of Multimodal Diffusion Models

• Akio Tomiya (Tokyo Woman's Christian U.) -- Lattice QCD with machine learning

• Kai Zhou (CUHK-SZ) -- Exploration of Matter in Extreme Conditions with Machine Learning

• Lingxiao Wang (RIKEN) -- Physics-driven Deep Learning and Generative Models

Speakers / Attendees

• Marc Bauer (Heidelberg U.)

• Léon Begiristain Ribo (Heidelberg U.)

• Tobias Buck (Heidelberg U.)

• Tiangang Cui (Heidelberg U.)

• Stefanie Czischek (U. Ottawa) -- Language Models for Large-Scale Quantum Many-Body Systems

• Wei-jie Fu (U. Dalian)

• Kenji Fukushima (U. Tokyo)

• Luigi Favaro (Heidelberg U.)

• Thomas Gasenzer (Heidelberg U.)

• Yuji Hirono (Osaka U.) -- Data-driven discovery of self-similarity using neural networks

• Jan Horak (Uniper) -- Algorithmic approaches to commodity trading

• Friederike Ihssen (Heidelberg U.) -- Physics-informed RG flows

• Renzo Kapust (Heidelberg U.) -- Super-Resolving Normalizing Flows for Lattice Field Theories

• Karina Koval (Heidelberg U.)

• Timoteo Lee (Heidelberg U.)

• Tatsuhiro Misumi (Kinki U.) -- Application of Physics-informed Neural Network (PINN) to Schroedinger-Newton equations

• Viktoria Noel (Heidelberg U.)

• Ayodele Ore (Heidelberg U.)

• Jan M. Pawlowski (Heidelberg U.)

• Yannic Pietschke (Heidelberg U.)

• Franz Sattler (Heidelberg U.)

• Tomas Schlenker (Heidelberg U.)

• Björn Malte Schäfer (Heidelberg U.)

• Manfred Salmhofer (Heidelberg U.)

• Ken Shiozaki (Yukawa Inst., Kyoto U.) -- A short introduction to non-Hermitian skin effect

• Dmitrij Sitenko (Heidelberg U.)

• Josephine Westermann (Heidelberg U.)

• Jojiro Yoshinaka (Kyoto U.) -- Neural Network Representation of Quantum Systems

• Jakob Zech (Heidelberg U.)