Zobrazeno 1 - 10
of 103
pro vyhledávání: '"Kahl, Karsten"'
Convolutional neural networks (CNNs) for image processing tend to focus on localized texture patterns, commonly referred to as texture bias. While most of the previous works in the literature focus on the task of image classification, we go beyond th
Externí odkaz:
http://arxiv.org/abs/2402.09530
Autor:
Ali, Ahsan, Brannick, James, Kahl, Karsten, Krzysik, Oliver A., Schroder, Jacob B., Southworth, Ben S.
Algebraic multigrid (AMG) is known to be an effective solver for many sparse symmetric positive definite (SPD) linear systems. For SPD systems, the convergence theory of AMG is well-understood in terms of the $A$-norm, but in a nonsymmetric setting,
Externí odkaz:
http://arxiv.org/abs/2401.11146
Obtaining annotations for complex computer vision tasks such as object detection is an expensive and time-intense endeavor involving a large number of human workers or expert opinions. Reducing the amount of annotations required while maintaining alg
Externí odkaz:
http://arxiv.org/abs/2310.00372
Autor:
Ali, Ahsan, Brannick, James, Kahl, Karsten, Krzysik, Oliver A., Schroder, Jacob B., Southworth, Ben S.
Publikováno v:
SIAM J. SCI. COMPUT. 2024 Vol. 0, No. 0, pp. S96-S122
This paper focuses on developing a reduction-based algebraic multigrid method that is suitable for solving general (non)symmetric linear systems and is naturally robust from pure advection to pure diffusion. Initial motivation comes from a new reduct
Externí odkaz:
http://arxiv.org/abs/2307.00229
Autor:
Riedlinger, Tobias, Schubert, Marius, Penquitt, Sarina, Kezmann, Jan-Marcel, Colling, Pascal, Kahl, Karsten, Roese-Koerner, Lutz, Arnold, Michael, Zimmermann, Urs, Rottmann, Matthias
Object detection on Lidar point cloud data is a promising technology for autonomous driving and robotics which has seen a significant rise in performance and accuracy during recent years. Particularly uncertainty estimation is a crucial component for
Externí odkaz:
http://arxiv.org/abs/2306.07835
Autor:
Schubert, Marius, Riedlinger, Tobias, Kahl, Karsten, Kröll, Daniel, Schoenen, Sebastian, Šegvić, Siniša, Rottmann, Matthias
Labeling datasets for supervised object detection is a dull and time-consuming task. Errors can be easily introduced during annotation and overlooked during review, yielding inaccurate benchmarks and performance degradation of deep neural networks tr
Externí odkaz:
http://arxiv.org/abs/2303.06999
For several classes of mathematical models that yield linear systems, the splitting of the matrix into its Hermitian and skew Hermitian parts is naturally related to properties of the underlying model. This is particularly so for discretizations of d
Externí odkaz:
http://arxiv.org/abs/2212.14208
Active learning as a paradigm in deep learning is especially important in applications involving intricate perception tasks such as object detection where labels are difficult and expensive to acquire. Development of active learning methods in such f
Externí odkaz:
http://arxiv.org/abs/2212.10836
Current state-of-the-art deep neural networks for image classification are made up of 10 - 100 million learnable weights and are therefore inherently prone to overfitting. The complexity of the weight count can be seen as a function of the number of
Externí odkaz:
http://arxiv.org/abs/2211.05525
A common way to approximate $F(A)b$ -- the action of a matrix function on a vector -- is to use the Arnoldi approximation. Since a new vector needs to be generated and stored in every iteration, one is often forced to rely on restart algorithms which
Externí odkaz:
http://arxiv.org/abs/2205.13842