Published on Tue Jun 16 2020

Preserving Dynamic Attention for Long-Term Spatial-Temporal Prediction

Haoxing Lin, Rufan Bai, Weijia Jia, Xinyu Yang, Yongjian You

Long-term predictions have been increasingly demanded in urban-wise data mining systems. We propose a Dynamic Switch-Attention Network (DSAN) with a novel Multi-Space Attention (MSA)mechanism that measures the correlations between inputs and outputs explicitly.

0
0
0
Abstract

Effective long-term predictions have been increasingly demanded in urban-wise data mining systems. Many practical applications, such as accident prevention and resource pre-allocation, require an extended period for preparation. However, challenges come as long-term prediction is highly error-sensitive, which becomes more critical when predicting urban-wise phenomena with complicated and dynamic spatial-temporal correlation. Specifically, since the amount of valuable correlation is limited, enormous irrelevant features introduce noises that trigger increased prediction errors. Besides, after each time step, the errors can traverse through the correlations and reach the spatial-temporal positions in every future prediction, leading to significant error propagation. To address these issues, we propose a Dynamic Switch-Attention Network (DSAN) with a novel Multi-Space Attention (MSA) mechanism that measures the correlations between inputs and outputs explicitly. To filter out irrelevant noises and alleviate the error propagation, DSAN dynamically extracts valuable information by applying self-attention over the noisy input and bridges each output directly to the purified inputs via implementing a switch-attention mechanism. Through extensive experiments on two spatial-temporal prediction tasks, we demonstrate the superior advantage of DSAN in both short-term and long-term predictions.

Fri Nov 29 2019
Machine Learning
ST-GRAT: A Novel Spatio-temporal Graph Attention Network for Accurately Forecasting Dynamically Changing Road Speed
Predicting road traffic speed is a challenging task due to different types of roadway and abrupt speed change. This paper proposes a novel Spatio-temporal graph attention that effectively captures the dynamics in road networks. The experimental results show that ST-GRAT outperforms existing models.
0
0
0
Fri Dec 13 2019
Artificial Intelligence
Spatial-Temporal Self-Attention Network for Flow Prediction
Flow prediction (e.g., crowd flow, traffic flow) is increasingly investigated in AI research field. It is very difficult due to complicated spatial dependencies between different locations and dynamic temporal dependencies among different time intervals. We propose a Spatial-Temporal Self-Attention Network (ST-
0
0
0
Thu Jan 09 2020
Machine Learning
Spatial-Temporal Transformer Networks for Traffic Flow Forecasting
Traffic forecasting has emerged as a core component of intelligent transportation systems. However, timely accurate traffic forecasting, especially long-term forecasting, still remains an open challenge due to the highly nonlinear and dynamic spatial-temporal dependencies of traffic flows. We propose a novel paradigm of Spatial-Temporal Transformer Networks.
0
0
0
Sat Feb 22 2020
Machine Learning
Interpretable Crowd Flow Prediction with Spatial-Temporal Self-Attention
Crowd flow prediction has been increasingly investigated in intelligent urban computing field. The most difficult part of predicting crowd flow is to measure the complicated spatial-temporal dependencies. We propose a Spatial-Temporal Self-Attention Network (STSAN) with an ST encoding gate.
0
0
0
Wed Aug 18 2021
Machine Learning
Multivariate and Propagation Graph Attention Network for Spatial-Temporal Prediction with Outdoor Cellular Traffic
Spatial-temporal prediction is a critical problem for intelligent transportation. Previous studies rely on large-scale traffic data collected from sensors. This paper addresses the problem via outdoor cellular traffic distilled from two billion records per day.
1
0
0
Mon Nov 11 2019
Machine Learning
GMAN: A Graph Multi-Attention Network for Traffic Prediction
Long-term traffic prediction is highly challenging due to the complexity of traffic systems. We propose a graph multi-attention network (GMAN) to predict traffic conditions for time steps ahead at different locations on a road network graph.
0
0
0