PulseAugur / Brief
EN
LIVE 08:18:24

Brief

last 24h
[2/2] 224 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

  1. A Coding Implementation on Spatial Graph Neural Networks for Urban Function Inference Using city2graph, OSMnx, and PyTorch Geometric

    This tutorial demonstrates how to build a spatial graph learning pipeline for urban function inference. It utilizes libraries like city2graph, OSMnx, and PyTorch Geometric to process OpenStreetMap data, construct graph structures, and train a GraphSAGE model. The process involves collecting Points of Interest (POI) and street network data, engineering spatial features, and creating both heterogeneous and homogeneous graph representations for predicting POI categories based on spatial context. AI

    A Coding Implementation on Spatial Graph Neural Networks for Urban Function Inference Using city2graph, OSMnx, and PyTorch Geometric

    IMPACT Provides a practical guide for applying graph neural networks to urban planning and analysis.

  2. On Efficient Scaling of GNNs via IO-Aware Layers Implementations

    Researchers have developed new GPU kernels to optimize Graph Neural Networks (GNNs) by addressing memory access bottlenecks. These kernels are designed to reduce data movement and improve locality for three main GNN layer families: SpMM-based convolutions, reduction-based aggregations, and attention-based layers. The implementations offer significant speedups, with some attention kernels achieving up to 8.5x faster performance and substantial memory reductions. AI

    IMPACT Optimized kernels could accelerate research and deployment of GNNs across various AI applications.