AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Graph-Based Semi-Supervised Learning with Non-ignorable Non-response

Fan Zhou, Tengfei Li, Haibo Zhou, Hongtu Zhu, Ye Jieping

NeurIPS 2019 · Graph learning › Graph self-supervised learning · approach: Heterophily and imbalance on graphs · 8 citations (OpenAlex)

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
model non-ignorable missingness (GNM) instead of standard GCN (SM)GNM · simulated data p=4, sigma=0.5RMSE: 1.1925 -> 0.6983Improved
model non-ignorable missingness (GNM) instead of standard GCN (SM)GNM · simulated data p=16, sigma=1RMSE: 1.4212 -> 1.1316Improved
model non-ignorable missingness (GNM) instead of standard GCN (SM)GNM · simulated data p=4, sigma=1MAPE: 0.3104 -> 0.2263Improved
use GNM instead of SMGNM · Cora binary simple setup lambda=1accuracy: 0.8683 -> 0.8947Improved
use GNM instead of SMGNM · Cora binary simple setup lambda=2accuracy: 0.8052 -> 0.8648Improved
use GNM instead of inverse-weighting method of RossetGNM · Cora binary simple setup lambda=1.5accuracy: 0.8311 -> 0.8908Improved
increase label imbalance lambda from 1 to 2 (SM)SM · Cora binary simple setupaccuracy: 0.8683 -> 0.8052Hurt
increase label imbalance lambda from 1 to 2 (GNM)GNM · Cora binary simple setupaccuracy: 0.8947 -> 0.8648Hurt
model non-ignorable missingness correctly instead of treating it as ignorable (GIM) (ablation)GNM · Cora binary complicated setup lambda=1accuracy: 0.8713 -> 0.8961Improved
model non-ignorable missingness correctly instead of treating it as ignorable (GIM) (ablation)GNM · Cora binary complicated setup lambda=2accuracy: 0.8291 -> 0.8669Improved
treat missingness as ignorable (GIM) instead of ignoring missingness altogether (SM) (ablation)GNM · Cora binary complicated setup lambda=2accuracy: 0.8141 -> 0.8291Improved
use GNM instead of Rosset in terms of varianceGNM · Cora binary simple setup lambda=1accuracy SD: 5.19e-2 -> 6.47e-3Improved

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