AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Fitting Autoregressive Graph Generative Models through Maximum Likelihood Estimation

Xu Han, Xiaohui Chen, Francisco J. R. Ruiz, Li-Ping Liu

JMLR 2023 · Graph learning › Graph self-supervised learning · approach: Diffusion-based graph generation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Rout-VI instead of ROS-VI inference networkGraphRNN · Community-smalltest log-likelihood (importance sampling): -53.7 -> -25.2Improved
use Rout-VI instead of ROS-VI inference networkGraphRNN · Enzymestest log-likelihood (importance sampling): -274.9 -> -92.6Improved
use a learned variational order distribution (Rout-VI) instead of uniform ordersGraphRNN · Community-smalltest log-likelihood (importance sampling): -154.6 -> -25.2Improved
use Rout-VI instead of uniform node ordersGraphDF · Citeseer-smalltest log-likelihood (importance sampling): -181.4 -> -93.5Improved
use Rout-VI instead of the DFS canonical orderGraphGEN · Community-smalltest log-likelihood (importance sampling): -263.74 -> -22.5Improved
use Rout-VI instead of ROS-VIGraphGEN · Enzymestest log-likelihood (importance sampling): -189.7 -> -89.5Improved
replace GraphGEN with the DAGG attention generative model, both trained with Rout-VIDAGG · Enzymestest log-likelihood (importance sampling): -89.5 -> -75.1Improved
use Rout-VI instead of DFS orderGraphGEN · Community-smallMMD Deg: 0.070 -> 0.094Hurt
use Rout-VI instead of BFS orderGraphRNN · CoraMMD Deg: 1.125 -> 0.049Improved
use Rout-VI instead of ROS-VI inference networkGraphRNN · Inference timetime to sample node orders: 20-times speed-upImproved
use Rout-VI instead of ROS-VIGraphRNN · Yeastlog-likelihood vs exact (importance sampling check)No clear effect
increase graph sizeDAGG and GraphRNN · Cora and Community-small subgraphsMMD Orbit and ClusterHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy