Scaling Properties of Deep Residual Networks
arXiv:2105.12245
Abstract
Residual networks (ResNets) have displayed impressive results in pattern recognition and, recently, have garnered considerable theoretical interest due to a perceived link with neural ordinary differential equations (neural ODEs). This link relies on the convergence of network weights to a smooth function as the number of layers increases. We investigate the properties of weights trained by stochastic gradient descent and their scaling with network depth through detailed numerical experiments. We observe the existence of scaling regimes markedly different from those assumed in neural ODE literature. Depending on certain features of the network architecture, such as the smoothness of the activation function, one may obtain an alternative ODE limit, a stochastic differential equation or neither of these. These findings cast doubts on the validity of the neural ODE model as an adequate asymptotic description of deep ResNets and point to an alternative class of differential equations as a better description of the deep network limit.
arXiv commentPublished at ICML 2021
References in corpus (4)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Analysis of the Gradient Descent Algorithm for a Deep Neural Network Model with Skip-connections
- Algorithm-Dependent Generalization Bounds for Overparameterized Deep Residual Networks
- ResNet After All? Neural ODEs and Their Numerical Solution