Vectorized Adjoint Sensitivity Method for Graph Convolutional Neural Ordinary Differential Equations
This work addresses a specific bottleneck for deploying GCDEs in resource-constrained hardware settings, offering an incremental improvement by vectorizing an existing method.
The paper tackles the problem of efficiently computing gradients for Graph Convolutional Neural Ordinary Differential Equations (GCDEs) in hardware-limited environments like edge computing and memristor crossbars, where autograd functions are unavailable, by providing a vectorized implementation of the adjoint sensitivity method.
This document, as the title stated, is meant to provide a vectorized implementation of adjoint dynamics calculation for Graph Convolutional Neural Ordinary Differential Equations (GCDE). The adjoint sensitivity method is the gradient approximation method for neural ODEs that replaces the back propagation. When implemented on libraries such as PyTorch or Tensorflow, the adjoint can be calculated by autograd functions without the need for a hand-derived formula. In applications such as edge computing and in memristor crossbars, however, autograds are not available, and therefore we need a vectorized derivation of adjoint dynamics to efficiently map the system on hardware. This document will go over the basics, then move on to derive the vectorized adjoint dynamics for GCDE.