Research & projects

Generative modelling, graph learning, and optimization. Explore the ideas through small, interactive synthetic experiments.

Encode once. Then diffuse.A-0.90B0.80C0.10D-0.30E0.65F-0.55A-0.87B0.83C0.15D-0.42E0.75F-0.68A-0.23B0.06C0.15D0.08E-0.04F-0.26Raw inputEncode onceDiffuse × 2→→ABCDEFSigned node featuresNode
ICAART 2024

DGDNN: Decoupled Graph Diffusion Neural Network for Stock Movement Prediction

Zinuo You, Zijian Shi, Hongbo Bo, John Cartlidge, Li Zhang, Yan Ge

Keep feature transformation separate from relational mixing over a dynamic stock graph.

Extended version · LNCS 2025 · ICAART 2024 revised selected papers: Multi-Stock Movements Classification: A Decoupled Graph Diffusion Neural Network Approach