Policy Message Passing: A New Algorithm for Probabilistic Graph Inference

TL;DR

Policy Message Passing introduces a probabilistic, stochastic approach to graph neural networks, enhancing reasoning capabilities and robustness, leading to superior performance on complex graph tasks.

Contribution

It presents a novel probabilistic algorithm for graph inference that leverages reasoning history and stochastic processes, outperforming existing models.

Findings

Consistently outperforms state-of-the-art models on complex graph tasks.

Utilizes reasoning history for improved inference accuracy.

Robust to noisy edges in graph data.

Abstract

A general graph-structured neural network architecture operates on graphs through two core components: (1) complex enough message functions; (2) a fixed information aggregation process. In this paper, we present the Policy Message Passing algorithm, which takes a probabilistic perspective and reformulates the whole information aggregation as stochastic sequential processes. The algorithm works on a much larger search space, utilizes reasoning history to perform inference, and is robust to noisy edges. We apply our algorithm to multiple complex graph reasoning and prediction tasks and show that our algorithm consistently outperforms state-of-the-art graph-structured models by a significant margin.