no code implementations • NeurIPS 2018 • Christoph Zimmer, Mona Meister, Duy Nguyen-Tuong
Learning time-series models is useful for many applications, such as simulation and forecasting.
1 code implementation • 16 Jun 2023 • Matthias Bitzer, Mona Meister, Christoph Zimmer
We propose amortizing kernel parameter inference over a complete kernel-structure-family rather than a fixed kernel structure.
1 code implementation • 17 Mar 2023 • Matthias Bitzer, Mona Meister, Christoph Zimmer
Machine learning methods that are used to produce the surrogate model should therefore address these problems by providing a scheme to keep the number of queries small, e. g. by using active learning and be able to capture the nonlinear and nonstationary properties of the system.
1 code implementation • 21 Oct 2022 • Matthias Bitzer, Mona Meister, Christoph Zimmer
Despite recent advances in automated machine learning, model selection is still a complex and computationally intensive process.
no code implementations • 1 Dec 2016 • Philipp Thomann, Ingrid Blaschzyk, Mona Meister, Ingo Steinwart
Our contributions are two fold: On the theoretical side we establish an oracle inequality for the overall learning method using the hinge loss, and show that the resulting rates match those known for SVMs solving the complete optimization problem with Gaussian kernels.