GitHub

Physics-Informed Graph Attention Network with Temporal Convolutional Encoder for Multi-Node Distribution Grid State Forecasting under Solar PV Uncertainty

There's not enough diversity for attention to learn meaningful patterns Uniform averaging is near-optimal when all neighbors matter equally Attention adds noise rather than extracting signal Despite ...
IEEE

MBGCN: Multi-view Block-wise Graph Convolutional Networks on Large-scale Graphs

Abstract: Existing methods based on graph convolutional network often struggle with large-scale graphs due to their high computing consumption and inefficiency. Although strategies such as edge ...
IEEE

Hyperspectral Image Classification via Dynamic Adaptive Graph Convolutional Network

Abstract: A dynamic graph convolutional network (DGCN) can represent temporal evolutionary features. Its compatibility with the spectral-dimensional characteristics of hyperspectral images (HSIs), ...
GitHub

cuda_graph_for_multi_modal_encoder.md

In multimodal reasoning services, the visual encoder (ViT / Vision Transformer) typically has a few characteristic traits: Many layers, fragmented operators: Each layer includes LN, QKV projections, ...