Probabilistic Approximation of Stochastic Time Series Using Bayesian Recurrent Neural Network
- 28 January 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (6) , 11664-11671
Abstract
In this brief, we investigate the approximation theory (AT) of Bayesian recurrent neural network (BRNN) for stochastic time series forecasting (TSF) from a probabilistic standpoint. Due to the cumulative dependencies present in stochastic time series, which are incompatible with the recurrent structure of BRNN and further complicate the analysis of AT, we first perform marginalization and transform the time series into a probabilistically equivalent latent variable model (LVM). Subsequently, we analyze the AT by evaluating the approximation error between the output mean of BRNN and that of the LVM, which are derived through Taylor expansion-based uncertainty propagation and distribution parameterization, respectively. Finally, leveraging the Khinchin’s law of large numbers, we study the convergence in probability of the sampling-based training algorithm, i.e., Bayes by Backprop (BBB), and prove that increasing the number of Monte Carlo samples in BBB leads to a convergence probability approaching one. Numerical simulations are conducted to demonstrate the validity of our results.Keywords
Funding Information
- National Key Research and Development Program of China (2021YFB2601700)
- Funds of the National Natural Science Foundation of China (U2133210)
- Fundamental Research Funds for the Central Universities (YWF-24-JT-102)
This publication has 24 references indexed in Scilit:
- Accurate medium-range global weather forecasting with 3D neural networksNature, 2023
- A Deep Learning Approach for Flight Delay Prediction Through Time-Evolving GraphsIEEE Transactions on Intelligent Transportation Systems, 2021
- EVDHM-ARIMA-Based Time Series Forecasting Model and Its Application for COVID-19 CasesIEEE Transactions on Instrumentation and Measurement, 2020
- Trajectory Generation by Chance-Constrained Nonlinear MPC With Probabilistic PredictionIEEE Transactions on Cybernetics, 2020
- A Fuzzy Interval Time-Series Energy and Financial Forecasting Model Using Network-Based Multiple Time-Frequency Spaces and the Induced-Ordered Weighted Averaging Aggregation OperationIEEE Transactions on Fuzzy Systems, 2020
- Probabilistic Forecasting With Fuzzy Time SeriesIEEE Transactions on Fuzzy Systems, 2019
- Adaptive Probabilistic Vehicle Trajectory Prediction Through Physically Feasible Bayesian Recurrent Neural Network2019
- Probability2019
- 4-D Flight Trajectory Tracking: A Receding Horizon Approach Integrating Feedback Linearization and Scenario OptimizationIEEE Transactions on Control Systems Technology, 2018
- A hybrid linear-neural model for time series forecastingIEEE Transactions on Neural Networks, 2000