GNN科研入門(mén)計(jì)劃第二期

快放假了,來(lái)學(xué)習(xí)吧~
GNN科研入門(mén)計(jì)劃第二期:圖神經(jīng)網(wǎng)絡(luò)理論、時(shí)間序列預(yù)測(cè)(時(shí)空模型)
面向?qū)ο螅篏NN科研入門(mén)的大三、大四、研一、博一學(xué)生
要求:掌握基本的Python、Pytorch語(yǔ)法?
形式:線上代碼講解+作業(yè)布置/講解+組會(huì)+研究方向討論
圖神經(jīng)網(wǎng)絡(luò)理論

周期:3個(gè)月,5次授課(全是直播形式),2h/次
價(jià)格:149元
具體內(nèi)容:
空間域卷積理論
譜域卷積理論
譜域卷與空間域卷積的關(guān)系
消息傳遞
GCN的底層實(shí)現(xiàn)(PyG版)
譜域?yàn)V波器設(shè)計(jì)
項(xiàng)目代碼講解一:GAT、GraphSAGE
項(xiàng)目代碼講解二:BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation
項(xiàng)目代碼講解三:LON-GNN: Spectral GNNs with Learnable Orthonormal Basis
2. 時(shí)間序列預(yù)測(cè)

周期:3個(gè)月,8次授課(2次錄播+6次直播),2h/次
價(jià)格:199元
注意:項(xiàng)目會(huì)有一定難度,都是最新的科研論文項(xiàng)目代碼講解
具體內(nèi)容:
?Graph WaveNet:?Graph WaveNet for Deep Spatial-Temporal Graph Modeling
?Ada-STNet:?Adaptive Spatio-temporal Graph Neural Network for traffic forecasting
AdapGL:??Adaptive Graph Spatial-Temporal Transformer Network for Traffic Flow Forecasting
TSAT:?Expressing Multivariate Time Series as Graphs with Time Series Attention
STGODE:?Spatial-Temporal Graph ODE Networks for Traffic Flow Forecasting
TLNets:?Transformation Learning Networks for long-range time-series prediction
GRAM-ODE:?Graph-based Multi-ODE Neural Networks for SpatioTemporal Traffic Forecasting
二選一
GraFITi:?Forecasting Irregularly Sampled Time Series using Graphs
STID: Spatial-Temporal Identity: A Simple yet Effective Baseline for Multivariate Time Series Forecasting