3 papers
math.DS2026
Architecture Induces Structural Invariant Manifolds of Neural Network Training Dynamics
Jiajie Zhao, Tao Luo, Yaoyu Zhang
While architecture is recognized as key to the performance of deep neural networks, its precise effect on training dynamics has been unclear due to the confounding influence of dat…
cs.LG2025
Embedding principle of homogeneous neural network for classification problem
Jiahan Zhang, Yaoyu Zhang, Tao Luo
In this paper, we study the Karush-Kuhn-Tucker (KKT) points of the associated maximum-margin problem in homogeneous neural networks, including fully-connected and convolutional neu…
cs.LG2025
Uncovering Critical Sets of Deep Neural Networks via Sample-Independent Critical Lifting
Leyang Zhang, Yaoyu Zhang, Tao Luo
This paper investigates the sample dependence of critical points for neural networks. We introduce a sample-independent critical lifting operator that associates a parameter of one…