关注
Zihan Wang
Zihan Wang
New York University
在 nyu.edu 的电子邮件经过验证
标题
引用次数
引用次数
年份
Reconstructing training data from model gradient, provably
Z Wang, J Lee, Q Lei
International Conference on Artificial Intelligence and Statistics, 6595-6612, 2023
212023
Implicit bias of SGD in -regularized linear DNNs: One-way jumps from high to low rank
Z Wang, A Jacot
arXiv preprint arXiv:2305.16038, 2023
132023
Data reconstruction attacks and defenses: A systematic evaluation
S Liu, Z Wang, Y Chen, Q Lei
arXiv preprint arXiv:2402.09478, 2024
42024
Wide neural networks trained with weight decay provably exhibit neural collapse
A Jacot, P Súkeník, Z Wang, M Mondelli
arXiv preprint arXiv:2410.04887, 2024
12024
系统目前无法执行此操作，请稍后再试。
文章 1–4