[Paper Review] Graph Normalizing Flows

Jenny Liu, Aviral Kumar+3 more|arXiv (Cornell University)|May 30, 2019

Advanced Graph Neural NetworksComputer Science30 references39 citations

TL;DR

The provided text is NeurIPS formatting instructions, not the actual paper content for Graph Normalizing Flows.

ABSTRACT

We introduce graph normalizing flows: a new, reversible graph neural network model for prediction and generation. On supervised tasks, graph normalizing flows perform similarly to message passing neural networks, but at a significantly reduced memory footprint, allowing them to scale to larger graphs. In the unsupervised case, we combine graph normalizing flows with a novel graph auto-encoder to create a generative model of graph structures. Our model is permutation-invariant, generating entire graphs with a single feed-forward pass, and achieves competitive results with the state-of-the art auto-regressive models, while being better suited to parallel computing architectures.

Motivation & Objective

• The input does not contain the paper's objectives; it appears to be NeurIPS 2019 formatting guidelines rather than the Graph Normalizing Flows content.

Proposed method

• No methodological details are available in the provided text; it consists of formatting rules and examples for NeurIPS submissions.

Experimental results

Research questions

• RQ1 No research questions are provided in the supplied text.

Key findings

• No findings are provided in the supplied text.