Get Started
Home
Topics
Search
Library
Research questionHow can reservoir-based graph neural networks aggregate multi-hop information without over-smoothing?Graph neural networks often need deeper message passing to capture distant relationships, but depth can make node representations indistinguishable. Reservoir-based variants offer iterative propagation but may not provide structured aggregation across multiple hops.
AI
Machine Learning
Neural and Evolutionary Computing
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.Reservoir-Based Graph Convolutional NetworksThe source presents RGC-Net, which combines reservoir dynamics with structured graph convolution for graph classification and graph generation, including dynamic brain-connectivity graphs. Its reported evidence covers convergence, classification, generation, and over-smoothing behavior in the evaluated experiments.research paper · Sep 4, 2026
Related questions
How can graph neural networks capture long-range interactions without oversmoothing or oversquashing on diverse large graphs?How can graph pooling preserve global-to-local topology while adapting granularity for graph classification?How can node-level over-smoothing diagnostics guide virtual-node selection to alleviate over-squashing in GNNs?How can representation-learning methods for dynamic heterogeneous graphs be compared across temporal granularities?