TY - GEN
T1 - LABEL PROPAGATION ACROSS GRAPHS
T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
AU - Bayer, Artun
AU - Chowdhury, Arindam
AU - Segarra, Santiago
N1 - Publisher Copyright:
© 2022 IEEE
PY - 2022
Y1 - 2022
N2 - Graph neural networks (GNNs) have achieved superior performance on node classification tasks in the last few years. Commonly, this is framed in a transductive semi-supervised learning setup wherein the entire graph - including the target nodes to be labeled - is available for training. Driven in part by scalability, recent works have focused on the inductive case where only the labeled portion of a graph is available for training. In this context, our current work considers a challenging inductive setting where a set of labeled graphs are available for training while the unlabeled target graph is completely separate, i.e., there are no connections between labeled and unlabeled nodes. Under the implicit assumption that the testing and training graphs come from similar distributions, our goal is to develop a labeling function that generalizes to unobserved connectivity structures. To that end, we employ a graph neural tangent kernel (GNTK) that corresponds to infinitely wide GNNs to find correspondences between nodes in different graphs based on both the topology and the node features. We augment the capabilities of the GNTK with residual connections and empirically illustrate its performance gains on standard benchmarks.
AB - Graph neural networks (GNNs) have achieved superior performance on node classification tasks in the last few years. Commonly, this is framed in a transductive semi-supervised learning setup wherein the entire graph - including the target nodes to be labeled - is available for training. Driven in part by scalability, recent works have focused on the inductive case where only the labeled portion of a graph is available for training. In this context, our current work considers a challenging inductive setting where a set of labeled graphs are available for training while the unlabeled target graph is completely separate, i.e., there are no connections between labeled and unlabeled nodes. Under the implicit assumption that the testing and training graphs come from similar distributions, our goal is to develop a labeling function that generalizes to unobserved connectivity structures. To that end, we employ a graph neural tangent kernel (GNTK) that corresponds to infinitely wide GNNs to find correspondences between nodes in different graphs based on both the topology and the node features. We augment the capabilities of the GNTK with residual connections and empirically illustrate its performance gains on standard benchmarks.
KW - graph neural network
KW - graph representation learning
KW - neural tangent kernel
KW - Node classification
UR - http://www.scopus.com/inward/record.url?scp=85131264351&partnerID=8YFLogxK
U2 - 10.1109/ICASSP43922.2022.9746838
DO - 10.1109/ICASSP43922.2022.9746838
M3 - Conference contribution
AN - SCOPUS:85131264351
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 5483
EP - 5487
BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 May 2022 through 27 May 2022
ER -