Timezone: »
Existing popular methods for semi-supervised learning with Graph Neural Networks (such as the Graph Convolutional Network) provably cannot learn a general class of neighborhood mixing relationships. To address this weakness, we propose a new model, MixHop, that can learn these relationships, including difference operators, by repeatedly mixing feature representations of neighbors at various distances. MixHop requires no additional memory or computational complexity, and outperforms on challenging baselines. In addition, we propose sparsity regularization that allows us to visualize how the network prioritizes neighborhood information across different graph datasets. Our analysis of the learned architectures reveals that neighborhood mixing varies per datasets.
Author Information
Sami Abu-El-Haija (USC Information Sciences Institute)
Bryan Perozzi (Google AI)
Amol Kapoor (Google Research)
Nazanin Alipourfard (University of Southern California)
Kristina Lerman (ISI, University of Southern California)
Hrayr Harutyunyan (University of Southern California)
Greg Ver Steeg (University of Southern California)
Aram Galstyan (USC ISI)
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Oral: MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing »
Thu. Jun 13th 12:05 -- 12:10 AM Room Grand Ballroom
More from the Same Authors
-
2023 Poster: Graph Generative Model for Benchmarking Graph Neural Networks »
Minji Yoon · Yue Wu · John Palowitch · Bryan Perozzi · Ruslan Salakhutdinov -
2022 : Robust GNNs Q/A »
Bryan Perozzi -
2022 : Robust GNNs »
Bryan Perozzi -
2022 : Challenges with Graph Neural Networks »
Bryan Perozzi -
2022 Expo Talk Panel: Challenges Of Applying Graph Neural Networks »
Bryan Perozzi · Vahab Mirrokni -
2020 Poster: Improving generalization by controlling label-noise information in neural network weights »
Hrayr Harutyunyan · Kyle Reing · Greg Ver Steeg · Aram Galstyan -
2020 Poster: All in the Exponential Family: Bregman Duality in Thermodynamic Variational Inference »
Rob Brekelmans · Vaden Masrani · Frank Wood · Greg Ver Steeg · Aram Galstyan