GAP: Generalizable Approximate Graph Partitioning Framework

TL;DR

GAP introduces a deep learning framework for graph partitioning that generalizes across unseen graphs, achieving competitive results and significantly faster performance than traditional methods.

Contribution

GAP is the first deep learning-based graph partitioning framework that generalizes to unseen graphs and optimizes partitioning directly through a differentiable loss.

Findings

GAP achieves competitive partition quality.

GAP is up to 100 times faster than baseline methods.

GAP generalizes well to various unseen graph structures.

Abstract

Graph partitioning is the problem of dividing the nodes of a graph into balanced partitions while minimizing the edge cut across the partitions. Due to its combinatorial nature, many approximate solutions have been developed, including variants of multi-level methods and spectral clustering. We propose GAP, a Generalizable Approximate Partitioning framework that takes a deep learning approach to graph partitioning. We define a differentiable loss function that represents the partitioning objective and use backpropagation to optimize the network parameters. Unlike baselines that redo the optimization per graph, GAP is capable of generalization, allowing us to train models that produce performant partitions at inference time, even on unseen graphs. Furthermore, because we learn the representation of the graph while jointly optimizing for the partitioning loss function, GAP can be easily tuned for a variety of graph structures. We evaluate the performance of GAP on graphs of varying sizes and structures, including graphs of widely used machine learning models (e.g., ResNet, VGG, and Inception-V3), scale-free graphs, and random graphs. We show that GAP achieves competitive partitions while being up to 100 times faster than the baseline and generalizes to unseen graphs.