dLSTM: a new approach for anomaly detection using deep learning with delayed prediction

RT @masahiro_sakai: dLSTM: a new approach for anomaly detection using deep learning with delayed prediction https://t.co/iL1d6SN5kl 前職同僚の真矢…

dLSTM: a new approach for anomaly detection using deep learning with delayed prediction https://t.co/iL1d6SN5kl 前職同僚の真矢さん達。LSTMの予測結果と実測値の誤差を用いた時系列異常検知で、複数モードの問題を扱うため、複数モデルの中で最も誤差が小さい予測値を、実測値の観測後に選ぶ。