Convolutional neural networks combined with Runge–Kutta methods

A convolutional neural network can be constructed using numerical methods for solving dynamical systems, since the forward pass of regarded as a trajectory system. However, existing models based on solvers cannot avoid iterations implicit methods, which makes inefficient at inference time. In this paper, we reinterpret pre-activation Residual Networks (ResNets) and their variants from systems view. We consider that Runge-Kutta are fused into training these models. Moreover, propose novel approach to constructing high-order in order achieve higher efficiency. Our proposed referred Convolutional Neural (RKCNNs). The RKCNNs evaluated multiple benchmark datasets. experimental results show vastly superior other system models: they accuracy with much fewer resources. They also expand family systems.