Autoreservoir computing for multistep ahead prediction based on the spatiotemporal information transformation

Pei Chen South China University of Technology Rui Liu South China University of Technology Kazuyuki Aihara The University of Tokyo Luonan Chen Chinese Academy of Sciences

Machine Learning mathscidoc:2103.41001

Nature Communications, 11, (4568), 2020.9
We develop an auto-reservoir computing framework, Auto-Reservoir Neural Network (ARNN), to efficiently and accurately make multi-step-ahead predictions based on a short-term high-dimensional time series. Different from traditional reservoir computing whose reservoir is an external dynamical system irrelevant to the target system, ARNN directly transforms the observed high-dimensional dynamics as its reservoir, which maps the high-dimensional/spatial data to the future temporal values of a target variable based on our spatiotemporal information (STI) transformation. Thus, the multi-step prediction of the target variable is achieved in an accurate and computationally efficient manner. ARNN is successfully applied to both representative models and real-world datasets, all of which show satisfactory performance in the multi-step-ahead prediction, even when the data are peturbed by noise and when the system is time-varying. Actually, such ARNN transformation equivalently expands the sample size and thus has great potential in practical applications in artificial intelligence and machine learning.
multi-step-ahead prediction; short-term time series; reservoir computing; delay embedding; high-dimensional data; low-dimensional manifold
[ Download ] [ 2021-03-23 10:33:26 uploaded by leoraymath ] [ 218 downloads ] [ 0 comments ]
@inproceedings{pei2020autoreservoir,
  title={Autoreservoir computing for multistep ahead prediction based on the spatiotemporal information transformation},
  author={Pei Chen, Rui Liu, Kazuyuki Aihara, and Luonan Chen},
  url={http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20210323103326056287747},
  booktitle={Nature Communications},
  volume={11},
  number={4568},
  year={2020},
}
Pei Chen, Rui Liu, Kazuyuki Aihara, and Luonan Chen. Autoreservoir computing for multistep ahead prediction based on the spatiotemporal information transformation. 2020. Vol. 11. In Nature Communications. http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20210323103326056287747.
Please log in for comment!
 
 
Contact us: office-iccm@tsinghua.edu.cn | Copyright Reserved