2 code implementations • 11 Apr 2024 • Marcel Hallgarten, Julian Zapata, Martin Stoll, Katrin Renz, Andreas Zell
We assess existing state-of-the-art planners on our benchmark and show that neither rule-based nor learning-based planners can safely navigate the interPlan scenarios.
1 code implementation • 1 Dec 2023 • Theresa Wagner, John W. Pearson, Martin Stoll
In this paper we consider the numerical solution to the soft-margin support vector machine optimization problem.
no code implementations • 10 Aug 2023 • Steffen Hagedorn, Marcel Hallgarten, Martin Stoll, Alexandru Condurache
We systematically review state-of-the-art deep learning-based prediction, planning, and integrated prediction and planning models.
no code implementations • 16 Feb 2023 • Kirandeep Kour, Sergey Dolgov, Peter Benner, Martin Stoll, Max Pfeffer
High-dimensional data in the form of tensors are challenging for kernel classification methods.
1 code implementation • 2 Dec 2022 • Edgar Ivan Sanchez Medina, Steffen Linke, Martin Stoll, Kai Sundmacher
The accurate prediction of physicochemical properties of chemical compounds in mixtures (such as the activity coefficient at infinite dilution $\gamma_{ij}^\infty$) is essential for developing novel and more sustainable chemical processes.
no code implementations • 15 May 2022 • Jan Blechschmidt, Jan-Frederik Pietschman, Tom-Christian Riemer, Martin Stoll, Max Winkler
In this paper we focus on comparing machine learning approaches for quantum graphs, which are metric graphs, i. e., graphs with dedicated edge lengths, and an associated differential operator.
1 code implementation • 19 Nov 2021 • Franziska Nestler, Martin Stoll, Theresa Wagner
We propose the use of an ANOVA kernel, where we construct several kernels based on lower-dimensional feature spaces for which we provide fast algorithms realizing the matrix-vector products.
no code implementations • 15 Sep 2021 • Anahita Iravanizad, Edgar Ivan Sanchez Medina, Martin Stoll
In recent years, graph neural networks (GNNs) have gained increasing popularity and have shown very promising results for data that are represented by graphs.
1 code implementation • 16 Apr 2021 • Dominik Alfke, Miriam Gondos, Lucile Peroche, Martin Stoll
Among the many time series learning tasks of great importance, we here focus on semi-supervised learning based on a graph representation of the data.
1 code implementation • 3 Aug 2020 • Dominik Alfke, Martin Stoll
Our method overcomes these issues by utilizing the pseudoinverse of the Laplacian.
1 code implementation • 10 Jul 2020 • Kai Bergermann, Martin Stoll, Toni Volkmer
We generalize a graph-based multiclass semi-supervised classification technique based on diffuse interface methods to multilayer graphs.
1 code implementation • 12 Feb 2020 • Kirandeep Kour, Sergey Dolgov, Martin Stoll, Peter Benner
An increasing amount of collected data are high-dimensional multi-way arrays (tensors), and it is crucial for efficient learning algorithms to exploit this tensorial structure as much as possible.
no code implementations • 17 Dec 2019 • Martin Stoll
Efficient numerical linear algebra is a core ingredient in many applications across almost all scientific and industrial disciplines.
no code implementations • 15 Jul 2019 • Dominik Garmatter, Margherita Porcelli, Francesco Rinaldi, Martin Stoll
Optimal control problems including partial differential equation (PDE) as well as integer constraints merge the combinatorial difficulties of integer programming and the challenges related to large-scale systems resulting from discretized PDEs.
Numerical Analysis Numerical Analysis Optimization and Control 65K05, 90C06, 90C11, 93C20, 90C51
2 code implementations • 24 May 2019 • Dominik Alfke, Martin Stoll
Graph Convolutional Networks (GCNs) have proven to be successful tools for semi-supervised learning on graph-based datasets.
1 code implementation • 7 Sep 2018 • Pedro Mercado, Jessica Bosch, Martin Stoll
Signed networks contain both positive and negative kinds of interactions like friendship and enmity.
no code implementations • 14 Aug 2018 • Dominik Alfke, Daniel Potts, Martin Stoll, Toni Volkmer
The graph Laplacian is a standard tool in data science, machine learning, and image processing.
no code implementations • 18 Nov 2016 • Jessica Bosch, Steffen Klamt, Martin Stoll
Additionally, we show that the diffuse interface method can be used for the segmentation of data coming from hypergraphs.