3つの要点 ✔️ 多変量時系列データ分析に特化した新たなGNNアーキテクチャを提案 ✔️ 従来のGNNに,変数の属性や,複数の時系列データ同士の時間的・空間的依存関係を捉えられる新たなモジュールを追加 ✔️ 4種類の評価用データセットで性能評価実験を行ったところ,4種中3種のデータセットでSoTAを達成 Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks written by Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, Xiaojun Chang, Chengqi Zhang (Submitted on 24 May 2020) Comments: Accepted by KDD 2020. Subjects: Mac
![データの点と点をつなぐ!?GNNを用いた新たな多次元時系列データ分析手法の提案!](https://cdn-ak-scissors.b.st-hatena.com/image/square/1e64e7dfdab5787fc7904fd9cc1cf53400d96ee6/height=288;version=1;width=512/https%3A%2F%2Faisholar.s3.ap-northeast-1.amazonaws.com%2Fmedia%2FMay2021%2Fgnn%25E3%2581%25AE%25E5%25A4%259A%25E6%25AC%25A1%25E5%2585%2583%25E6%2599%2582%25E7%25B3%25BB%25E5%2588%2597%25E3%2583%2586%25E3%2582%2599%25E3%2583%25BC%25E3%2582%25BF%25E5%2588%2586%25E6%259E%2590%25E6%2589%258B%25E6%25B3%25951-min.png)