עקוב אחר
Florian Rossmannek
Florian Rossmannek
ETH Zurich, Department of Mathematics
כתובת אימייל מאומתת בדומיין math.ethz.ch - דף הבית
כותרת
צוטט על ידי
צוטט על ידי
שנה
Efficient approximation of high-dimensional functions with neural networks
P Cheridito, A Jentzen, F Rossmannek
IEEE Transactions on Neural Networks and Learning Systems 33 (7), 3079-3093, 2021
45*2021
Non-convergence of stochastic gradient descent in the training of deep neural networks
P Cheridito, A Jentzen, F Rossmannek
Journal of Complexity 64, 101540, 2021
382021
A proof of convergence for gradient descent in the training of artificial neural networks for constant target functions
P Cheridito, A Jentzen, A Riekert, F Rossmannek
Journal of Complexity 72, 101646, 2022
252022
Landscape analysis for shallow neural networks: complete classification of critical points for affine target functions
P Cheridito, A Jentzen, F Rossmannek
Journal of Nonlinear Science 32 (5), 64, 2022
182022
Gradient descent provably escapes saddle points in the training of shallow ReLU networks
P Cheridito, A Jentzen, F Rossmannek
arXiv preprint arXiv:2208.02083, 2022
42022
Efficient Sobolev approximation of linear parabolic PDEs in high dimensions
P Cheridito, F Rossmannek
arXiv preprint arXiv:2306.16811, 2023
32023
The curse of dimensionality and gradient-based training of neural networks: shrinking the gap between theory and applications
F Rossmannek
ETH Zurich, 2023
2023
המערכת אינה יכולה לבצע את הפעולה כעת. נסה שוב מאוחר יותר.
מאמרים 1–7