Følg
Mo Zhou
Mo Zhou
Verifisert e-postadresse på cs.washington.edu - Startside
Tittel
Sitert av
Sitert av
År
Toward Understanding the Importance of Noise in Training Neural Networks
M Zhou, T Liu, Y Li, D Lin, E Zhou, T Zhao
International Conference on Machine Learning, 7594-7602, 2019
972019
Towards understanding the importance of shortcut connections in residual networks
T Liu, M Chen, M Zhou, SS Du, E Zhou, T Zhao
Advances in Neural Information Processing Systems, 7892-7902, 2019
662019
A local convergence theory for mildly over-parameterized two-layer neural network
M Zhou, R Ge, C Jin
Conference on Learning Theory, 4577-4632, 2021
442021
Understanding Edge-of-Stability Training Dynamics with a Minimalist Example
X Zhu, Z Wang, X Wang, M Zhou, R Ge
International Conference on Learning Representations, 2023
392023
Understanding Deflation Process in Over-parametrized Tensor Decomposition
R Ge, Y Ren, X Wang, M Zhou
Advances in Neural Information Processing Systems 34, 1299-1311, 2021
232021
Plateau in Monotonic Linear Interpolation--A" Biased" View of Loss Landscape for Deep Networks
X Wang, AN Wang, M Zhou, R Ge
International Conference on Learning Representations, 2023
82023
Depth Separation with Multilayer Mean-Field Networks
Y Ren, M Zhou, R Ge
International Conference on Learning Representations, 2023
62023
Understanding The Robustness of Self-supervised Learning Through Topic Modeling
Z Luo, S Wu, C Weng, M Zhou, R Ge
International Conference on Learning Representations, 2023
6*2023
Implicit Regularization Leads to Benign Overfitting for Sparse Linear Regression
M Zhou, R Ge
International Conference on Machine Learning, 2023
22023
How Does Gradient Descent Learn Features--A Local Analysis for Regularized Two-Layer Neural Networks
M Zhou, R Ge
arXiv preprint arXiv:2406.01766, 2024
2024
Multi-head CLIP: Improving CLIP with diverse representations and flat minima
M Zhou, X Zhou, E Li, S Ermon, R Ge
NeurIPS 2023 OPT Workshop, 2023
2023
Systemet kan ikke utføre handlingen. Prøv på nytt senere.
Artikler 1–11