Convergent Graph Solvers

by   Junyoung Park, et al.

We propose the convergent graph solver (CGS), a deep learning method that learns iterative mappings to predict the properties of a graph system at its stationary state (fixed point) with guaranteed convergence. CGS systematically computes the fixed points of a target graph system and decodes them to estimate the stationary properties of the system without the prior knowledge of existing solvers or intermediate solutions. The forward propagation of CGS proceeds in three steps: (1) constructing the input dependent linear contracting iterative maps, (2) computing the fixed-points of the linear maps, and (3) decoding the fixed-points to estimate the properties. The contractivity of the constructed linear maps guarantees the existence and uniqueness of the fixed points following the Banach fixed point theorem. To train CGS efficiently, we also derive a tractable analytical expression for its gradient by leveraging the implicit function theorem. We evaluate the performance of CGS by applying it to various network-analytic and graph benchmark problems. The results indicate that CGS has competitive capabilities for predicting the stationary properties of graph systems, irrespective of whether the target systems are linear or non-linear. CGS also shows high performance for graph classification problems where the existence or the meaning of a fixed point is hard to be clearly defined, which highlights the potential of CGS as a general graph neural network architecture.


A Converse to Banach's Fixed Point Theorem and its CLS Completeness

Banach's fixed point theorem for contraction maps has been widely used t...

Iterative Neural Networks with Bounded Weights

A recent analysis of a model of iterative neural network in Hilbert spac...

Fixed-point iterative linear inverse solver with extended precision

Solving linear systems is a ubiquitous task in science and engineering. ...

Connections between spectral properties of asymptotic mappings and solutions to wireless network problems

In this study we establish connections between asymptotic functions and ...

Beyond Initial Algebras and Final Coalgebras

We provide a construction of the fixed points of functors which may not ...

On algorithmically boosting fixed-point computations

This paper is a thought experiment on exponentiating algorithms. One of ...

Fixed Point Networks: Implicit Depth Models with Jacobian-Free Backprop

A growing trend in deep learning replaces fixed depth models by approxim...

Please sign up or login with your details

Forgot password? Click here to reset