TY - GEN
T1 - Dyat nets
T2 - 28th International Joint Conference on Artificial Intelligence, IJCAI 2019
AU - Muralidhar, Nikhil
AU - Muthiah, Sathappah
AU - Ramakrishnan, Naren
N1 - Publisher Copyright:
© 2019 International Joint Conferences on Artificial Intelligence. All rights reserved.
PY - 2019
Y1 - 2019
N2 - Multivariate time series forecasting is an important task in state forecasting for cyber-physical systems (CPS). State forecasting in CPS is imperative for optimal planning of system energy utility and understanding normal operational characteristics of the system thus enabling anomaly detection. Forecasting models can also be used to identify sub-optimal or worn out components and are thereby useful for overall system monitoring. Most existing work only performs single step forecasting but in CPS it is imperative to forecast the next sequence of system states (i.e curve forecasting). In this paper, we propose DyAt (Dynamic Attention) networks, a novel deep learning sequence to sequence (Seq2Seq) model with a novel hierarchical attention mechanism for long-term time series state forecasting. We evaluate our method on several CPS state forecasting and electric load forecasting tasks and find that our proposed DyAt models yield a performance improvement of at least 13.69% for the CPS state forecasting task and a performance improvement of at least 18.83% for the electric load forecasting task over other state-of-the-art forecasting baselines. We perform rigorous experimentation with several variants of the DyAt model and demonstrate that the DyAt models indeed learn better representations over the entire course of the long term forecast as compared to their counterparts with or without traditional attention mechanisms. All data and source code has been made available online.
AB - Multivariate time series forecasting is an important task in state forecasting for cyber-physical systems (CPS). State forecasting in CPS is imperative for optimal planning of system energy utility and understanding normal operational characteristics of the system thus enabling anomaly detection. Forecasting models can also be used to identify sub-optimal or worn out components and are thereby useful for overall system monitoring. Most existing work only performs single step forecasting but in CPS it is imperative to forecast the next sequence of system states (i.e curve forecasting). In this paper, we propose DyAt (Dynamic Attention) networks, a novel deep learning sequence to sequence (Seq2Seq) model with a novel hierarchical attention mechanism for long-term time series state forecasting. We evaluate our method on several CPS state forecasting and electric load forecasting tasks and find that our proposed DyAt models yield a performance improvement of at least 13.69% for the CPS state forecasting task and a performance improvement of at least 18.83% for the electric load forecasting task over other state-of-the-art forecasting baselines. We perform rigorous experimentation with several variants of the DyAt model and demonstrate that the DyAt models indeed learn better representations over the entire course of the long term forecast as compared to their counterparts with or without traditional attention mechanisms. All data and source code has been made available online.
UR - http://www.scopus.com/inward/record.url?scp=85074922916&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85074922916&partnerID=8YFLogxK
U2 - 10.24963/ijcai.2019/441
DO - 10.24963/ijcai.2019/441
M3 - Conference contribution
AN - SCOPUS:85074922916
T3 - IJCAI International Joint Conference on Artificial Intelligence
SP - 3180
EP - 3186
BT - Proceedings of the 28th International Joint Conference on Artificial Intelligence, IJCAI 2019
A2 - Kraus, Sarit
Y2 - 10 August 2019 through 16 August 2019
ER -