Deep Probabilistic Time Series Forecasting using Augmented Recurrent Input for Dynamic Systems

TL;DR

This paper introduces a novel deep probabilistic sequence model combining deep generative models and state space models, using an augmented recurrent input space to improve dynamic system time series forecasting.

Contribution

It proposes a new RNN-based variational sequence model with lagged hybrid outputs and a generalized auto-regressive strategy for better dynamic pattern mining and training-prediction consistency.

Findings

Outperforms existing deep probabilistic SSM models on eight benchmarks.

Demonstrates superior accuracy in system identification tasks.

Effectively quantifies predictive distributions in real-world applications.

Abstract

The demand of probabilistic time series forecasting has been recently raised in various dynamic system scenarios, for example, system identification and prognostic and health management of machines. To this end, we combine the advances in both deep generative models and state space model (SSM) to come up with a novel, data-driven deep probabilistic sequence model. Specifically, we follow the popular encoder-decoder generative structure to build the recurrent neural networks (RNN) assisted variational sequence model on an augmented recurrent input space, which could induce rich stochastic sequence dependency. Besides, in order to alleviate the inconsistency issue of the posterior between training and predicting as well as improving the mining of dynamic patterns, we (i) propose using a lagged hybrid output as input for the posterior at next time step, which brings training and predicting into alignment; and (ii) further devise a generalized auto-regressive strategy that encodes all the historical dependencies for the posterior. Thereafter, we first investigate the methodological characteristics of the proposed deep probabilistic sequence model on toy cases, and then comprehensively demonstrate the superiority of our model against existing deep probabilistic SSM models through extensive numerical experiments on eight system identification benchmarks from various dynamic systems. Finally, we apply our sequence model to a real-world centrifugal compressor forecasting problem, and again verify its outstanding performance by quantifying the time series predictive distribution.