RNN with Particle Flow for Probabilistic Spatio-Temporal Forecasting
Spatio-temporal forecasting has numerous applications in analyzing wireless, traffic, and financial networks.
Many classical statistical models often fall short in handling the complexity and high non-linearity present in time-series data.
Recent advances in deep learning allow for better modelling of spatial and temporal dependencies.
While most of these models focus on obtaining accurate point forecasts, they do not characterize the prediction uncertainty.
In this work, we consider the time-series data as a random realization from a nonlinear state-space model and target Bayesian inference of the hidden states for probabilistic forecasting.
We use particle flow as the tool for approximating the posterior distribution of the states, as it is shown to be highly effective in complex, high-dimensional settings.
Thorough experimentation on several real world time-series datasets demonstrates that our approach provides better characterization of uncertainty while main taining comparable accuracy to the state-of-the-art point forecasting methods.
Authors:
- Soumyasundar Pal
- Florence Regol
- Liheng Ma
- Yingxue Zhang
- Mark Coates (Prof. McGill University)
Citation
This project was published at ICML 2020.
@inproceedings{pal2021,
author={S. Pal and L. Ma and Y. Zhang and M. Coates},
booktitle={Proc. Int. Conf. Machine Learning},
title={{RNN} with Particle Flow for Probabilistic Spatio-temporal Forecasting},
month={Jul.},
year={2021},
address={Virtual Conference}}