Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks
This work addresses the theoretical foundations of deep learning for researchers, offering incremental improvements in analyzing optimization and generalization in overparameterized networks.
The paper tackles the problem of understanding why overparameterized two-layer neural networks can fit data and generalize, providing a tighter characterization of training speed with random labels and a generalization bound independent of network size, with experiments showing clear distinctions on MNIST and CIFAR.
Recent works have cast some light on the mystery of why deep nets fit any data and generalize despite being very overparametrized. This paper analyzes training and generalization for a simple 2-layer ReLU net with random initialization, and provides the following improvements over recent works: (i) Using a tighter characterization of training speed than recent papers, an explanation for why training a neural net with random labels leads to slower training, as originally observed in [Zhang et al. ICLR'17]. (ii) Generalization bound independent of network size, using a data-dependent complexity measure. Our measure distinguishes clearly between random labels and true labels on MNIST and CIFAR, as shown by experiments. Moreover, recent papers require sample complexity to increase (slowly) with the size, while our sample complexity is completely independent of the network size. (iii) Learnability of a broad class of smooth functions by 2-layer ReLU nets trained via gradient descent. The key idea is to track dynamics of training and generalization via properties of a related kernel.