About Re-producing the dynamics of ResNet and i-ResNet in Figure 1

[Zihan-Liu-00]

Hi,

I'm trying to reproduce the phenomenon in Figure 1. However I got some confusion. As you demonstrated, networks in Fig. 1 map the interval [-2,2] to noisy x^3. As [-2,2] -> x^3 has merely one dimension, ResNets require 3 dimensions input. I wonder how they map that interval. If the dynamics are mapping the outputs of residual blocks, however, the outputs have different sizes due to the downsampling. In brief, my question is about how you did that mapping operation.

Mant thanks,
Z. L