GNN, GCNの手法がたくさん提案されている状況だが精度が頭打ち感
● RNN系:YujiaLi, Daniel Tarlow, Gated graph
sequence neural networks: ICLR2016
● CNN系:
○ William L Hamilton, Rex Ying, and Jure
Leskovec., Inductive representation
learning on large graphs: NIPS2017
○ Thomas N Kipf and Max Welling. Semi-
supervised classification with graph
convolutional networks: ICLR2017
● Attention系:Petar Velickovic, Guillem Cucurull,
Arantxa Casanova, Adriana Romero, Pietro Lio,
and Yoshua Bengio. Graph attention networks:
ICLR2018
● 無理やりDeepにしてみたよ系:Keyulu Xu,
Chengtao Li, Yonglong Tian, Tomohiro Sonobe,
Ken-ichi Kawarabayashi, and Stefanie Jegelka.
Representation learning on graphs with jumping
knowledge networks. ICML2018
2. 背景
Keyulu Xu, Chengtao Li, Yonglong Tian, Tomohiro Sonobe, Ken-ichi Kawarabayashi, and Stefanie
Jegelka. Representation learning on graphs with jumping knowledge networks. ICML2018
R-GCN GAT JK-Net
6.
Graph Neural NetworkはDeepにしても全然精度があがらない
2.背景
Kipf, Thomas N., and Max Welling. "Semi-supervised classification with
graph convolutional networks." arXiv preprint arXiv:1609.02907 (2016).