Our paper "On the universal transformation of data driven models to control systems" (Automatica) has just been selected for the Editor's choice award in March 2023. Follow the link to check out Editor in Chief Andrew Teel's comment on our work.
Our paper "Efficient Time-Stepping for Numerical Integration Using Reinforcement Learning" (DOI: 10.1137/21M1412682) has just appeared in the SIAM Journal on Scientific Computing
Our paper "Towards reliable data-based optimal and predictive control using extended DMD" (DOI: 10.1016/j.ifacol.2023.02.029) has just appeared in the proceedings of the 12th IFAC Symposium on Nonlinear Control Systems (NOLCOS)
We have just published a preprint where we try to make a strong point for the usage of dynamical models when using reinforcement learning (RL) for feedback control of dynamical systems governed by partial differential equations (PDEs). To breach the gap between the immense promises we see in RL and the applicability in complex engineering systems, the main challenges are the massive requirements in terms of the training data, as well as the lack…
We have just published a preprint where we present a convolutional framework which significantly reduces the complexity and thus, the computational effort for distributed reinforcement learning control of dynamical systems governed by partial differential equations (PDEs). Exploiting translational invariances, the high-dimensional distributed control problem can be transformed into a multi-agent control problem with many identical, uncoupled…
We have just published a preprint where we study the data-driven approximation of the Koopman operator for stochastic differential equations on reproducing kernel Hilbert spaces (RKHS). Our focus is on the estimation error if the data are collected from long-term ergodic simulations. We derive both an exact expression for the variance of the kernel cross-covariance operator, measured in the Hilbert-Schmidt norm, and probabilistic bounds for the…
Our paper "On the universal transformation of data-driven models to control systems" (DOI: 10.1016/j.automatica.2022.110840) has just appeared as an Open Access paper in the Journal Automatica.
Our paper "Finite-Data Error Bounds for Koopman-Based Prediction and Control" (DOI: 10.1007/s00332-022-09862-1) has just appeared as an Open Access paper in the Journal of Nonlinear Science.
Today, the start of a new AI Junior Research Group on Multicriteria Machine Learning (MultiML), was officially anounced. The aim of this group, which is part of the Data Science for Engineering group, is to develop multi-objective optimisation methods in order to make the training process of deep neural networks more robust, efficient and interactive and thus significantly improve it. Furthermore, the additional incorporation of system knowledge…
Our paper "On the structure of regularization paths for piecewise differentiable regularization terms" (DOI: 10.1007/s10898-022-01223-2) has appeared in the Journal of Global Optimization.
We have just published a preprint on a "Multi-Objective Trust-Region Filter Method for Nonlinear Constraints using Inexact Gradients". In this article, we build on previous work to present an optimization algorithm for nonlinearly constrained multi-objective optimization problems. The algorithm combines a surrogate-assisted derivative-free trust-region approach with the filter method known from single-objective optimization. Instead of the true…
We have just published a preprint where we derive efficient algorithms to compute weakly Pareto optimal solutions for smooth, convex and unconstrained multiobjective optimization problems in general Hilbert spaces. We employ Nesterov acceleration to define an algorithm with an improved convergence rate compared to the plain multiobjective steepest descent method (Algorithm 1). A further improvement in terms of efficiency is achieved by avoiding…
Our paper "Koopman analysis of quantum systems" (DOI: 10.1088/1751-8121/ac7d22) has appeared in the Journal of Physics A: Mathematical and Theoretical.
In October 2021, a new research project was initialized under the title "DARE: Training, validation and benchmark tools for the development of data-driven operation and control strategies for intelligent, local energy systems" (German: "DARE: Trainings-, Validierungs- und Benchmarkwerkzeuge zur Entwicklung datengetriebener Betriebs- und Regelungsverfahren für intelligente, lokale Energiesysteme")
Within the next two years, researchers from the S…
