Search

Time series forecasting is ubiquitous in various scientific and industrial domains. Powered by recurrent and convolutional and self-attention mechanism, deep learning exhibits high efficacy in time series forecasting. However, the existing forecasting methods are suffering some limitations. For example, recurrent neural networks are limited by the gradient vanishing problem, convolutional neural networks cost more parameters, and self-attention has a defect in capturing local dependencies. What's more, they all rely on time invariant or stationary since they leverage parameter sharing by repeating a set of fixed architectures with fixed parameters over time or space. To address the above issues, in this paper we propose a novel time-variant framework named Self-Attention-based Time-Variant Neural Networks (SATVNN), generally capable of capturing dynamic changes of time series on different scales more accurately with its time-variant structure and consisting of self-attention blocks that seek to better capture the dynamic changes of recent data, with the help of Gaussian distribution, Laplace distribution and a novel Cauchy distribution, respectively. SATVNN obviously outperforms the classical time series prediction methods and the state-of-the-art deep learning models on lots of widely used real-world datasets. [ABSTRACT FROM AUTHOR]

Correlated time series refer to multiple time series which are recorded simultaneously to monitor the changing of multiple observations in a whole system. Correlated time series prediction plays a significant role in many real-world applications to help people make reasonable decisions. Yet it is very challenging, because different from single time series, correlated time series show both intra-sequence temporal dependencies and inter-sequence spatial dependencies. In addition, correlated time series are also affected by external factors in actual scenarios. Although RNNs have been proved to be effective on sequential data modeling, existing related works only focus on sequential patterns in a single time series, failing to comprehensively consider the inter-dependencies among multiple time series, which is essential for correlated time series prediction. In this paper, we propose a novel variant of LSTM, named CTS-LSTM, to collectively forecast correlated time series. Specifically, spatial and temporal correlations are explicitly modeled and respectively maintained in cells to capture the complex non-linear patterns in correlated time series. A general interface for handling external factors is further designed to enhance forecasting performance of the model. Experiments are conducted on two types of real-world datasets, viz., civil aviation passenger demand data and air quality data. And our CTS-LSTM achieves at least 9.0%, 16.5% and 21.3% lower RMSE, MAE and MAPE compared to the state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

Accurate forecasting of time series mitigates the uncertainty of future outlooks and is a great help in reducing errors in decisions. Despite years of researches, there are still some challenges to accurate forecasting of time series, including the difficulty of dynamic modeling, the problem of capturing short-term correlations, and the conundrum of long-term forecasting. This paper offers an A dversarial Truncated Cauchy S elf- A ttentive T ime- V ariant Neural N etwork (ASATVN) for multi-step ahead time series forecasting. Specifically, the proposed model builds on Generative Adversarial Networks, in which the generator is composed of a novel time-variant model. The time-variant model contributes to learning dynamic time-series changes with its time-variant architecture and employs a newly proposed Truncated Cauchy Self-Attention block to capture the local sequential dependencies better. For the discriminator, two self-attentive discriminators are presented to regularize predictions with fidelity and continuity, which is beneficial to predicting sequence over longer time horizons. Our proposed ASATVN model outperforms the state-of-the-art predictive models on eleven real-world benchmark datasets, demonstrating its effectiveness. • A time-variant network learns various dynamics across multiple time steps. • A new self-attention are more sensitive to the local context of time series. • Two discriminators regularize predictor to offer realistic and continuous forecasts. [ABSTRACT FROM AUTHOR]