Neural network training via persistence of excitation Active learning for stochastic economic model predictive control

Neural ODE training via stochastic control and uncertainty quantification Benchmarks for active learning in systems and control

Active learning in the Koopman framework

Safe active learning control with Gaussian processes

ALeSCo's Research Projects

Illustration of applications

Illustration of applications

Illustration of applications

ALeSCo consists of six individual research projects that cover topics from data-informativity over modeling and control with neural networks, Gaussian Processes, and evolution operators to numerical optimization and benchmarking. It is aimed at exploring commonalities, synergies, and developing algorithms with the whole control pipeline in mind.

Individual research projects

Neural network training via persistence of excitation

Neural ODE training via stochastic control and uncertainty quantification

Active learning in the Koopman framework

Safe active learning control with Gaussian processes

Active learning for stochastic economic model predictive control

Benchmarks for active learning in systems and control

Data-driven feedback loop

Data-driven feedback loop

Data-driven feedback loop

Picture of ALeSCo principle investigators

Picture of ALeSCo principle investigators

Picture of ALeSCo principle investigators

Picture of the IRT (LUH)

Picture of the IRT (LUH)

Picture of the IRT (LUH)

Last Change: 27.05.26 Print