Quiver neural networks
From MaRDI portal
Publication:6406087
arXiv2207.12773MaRDI QIDQ6406087FDOQ6406087
Authors: Iordan Ganev, Robin Walters 
Publication date: 26 July 2022
Abstract: We develop a uniform theoretical approach towards the analysis of various neural network connectivity architectures by introducing the notion of a quiver neural network. Inspired by quiver representation theory in mathematics, this approach gives a compact way to capture elaborate data flows in complex network architectures. As an application, we use parameter space symmetries to prove a lossless model compression algorithm for quiver neural networks with certain non-pointwise activations known as rescaling activations. In the case of radial rescaling activations, we prove that training the compressed model with gradient descent is equivalent to training the original model with projected gradient descent.
This page was built for publication: Quiver neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6406087)
