collaborators

7 papers

cs.LG2025

Energy Matching: Unifying Flow Matching and Energy-Based Models for Generative Modeling

Michal Balcerak, Tamaz Amiranashvili, Antonio Terpin +5

Current state-of-the-art generative models map noise to data distributions by matching flows or scores. A key limitation of these models is their inability to readily integrate ava…

cs.NE2025

Data-Driven Discovery of Interpretable Kalman Filter Variants through Large Language Models and Genetic Programming

Vasileios Saketos, Sebastian Kaltenbach, Sergey Litvinov +1

Algorithmic discovery has traditionally relied on human ingenuity and extensive experimentation. Here we investigate whether a prominent scientific computing algorithm, the Kalman…

cs.LG2025

Reinforcement Learning Closures for Underresolved Partial Differential Equations using Synthetic Data

Lothar Heimbach, Sebastian Kaltenbach, Petr Karnakov +2

Partial Differential Equations (PDEs) describe phenomena ranging from turbulence and epidemics to quantum mechanics and financial markets. Despite recent advances in computational…

physics.flu-dyn2025

Optimal Lattice Boltzmann Closures through Multi-Agent Reinforcement Learning

Paul Fischer, Sebastian Kaltenbach, Sergey Litvinov +2

The Lattice Boltzmann method (LBM) offers a powerful and versatile approach to simulating diverse hydrodynamic phenomena, spanning microfluidics to aerodynamics. The vast range of…

cs.LG2025

Learning Effective Dynamics across Spatio-Temporal Scales of Complex Flows

Han Gao, Sebastian Kaltenbach, Petros Koumoutsakos

Modeling and simulation of complex fluid flows with dynamics that span multiple spatio-temporal scales is a fundamental challenge in many scientific and engineering domains. Full-s…

cs.LG2025

Closure Discovery for Coarse-Grained Partial Differential Equations Using Grid-based Reinforcement Learning

Jan-Philipp von Bassewitz, Sebastian Kaltenbach, Petros Koumoutsakos

Reliable predictions of critical phenomena, such as weather, wildfires and epidemics often rely on models described by Partial Differential Equations (PDEs). However, simulations t…