Deep Architecture for Traffic Flow Prediction: Deep Belief Networks With Multitask Learning

By Wenhao Huang, Guojie Song, Haikun Hong, Kunqing Xie. IEEE transactions on intelligent transportation systems (Print), 2014

1,046 citations recorded by the source corpus.

Subfields: Building and Construction, Civil and Structural Engineering, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/32f75852c5eb5ec6d254818d054e57a90e76b3ba

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Building and Construction

In Civil and Structural Engineering

In Control and Systems Engineering

Citations in the corpus

High-Order Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting (2018, background): “Deep learning models for traffic forecasting, like deep belief networks (DBN) [8] and stacked autoencoders [13], can effectively learn high dimensional features and achieve good forecasting performance.”
Modeling of nonlinear system based on deep learning framework (2015, background): “RBMs [10,11] can be interpreted as neural network models which consist of two types of units called visible neurons and hidden neurons.”
Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction (2017, methodology): “Deep learning methods are widely used in the domain of image processing with satisfactory results, since deep learning architectures usually have deeper construction and depict more complex nonlinear functions than other neural networks [25,27,30,39].”
Spectral–Spatial Classification of Hyperspectral Data Based on Deep Belief Network (2015, neutral): “with as less prior knowledge as possible [49].”
GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks (2017, background): “…processing (Hashimoto et al. (2016); Collobert & We-ston (2008); Søgaard & Goldberg (2016)) to speech synthesis (Wu et al. (2015); Seltzer & Droppo (2013)), from very domain-speciﬁc applications like trafﬁc prediction (Huang et al. (2014)) to very general cross-domain work (Bilen & Vedaldi (2017)).”
Survey on Deep Neural Networks in Speech and Vision Systems (2019, methodology): “However, recent advancements in IS are permeating almost every aspect of our life with the introduction of intelligent transportation [1-10], intelligent diagnosis and health monitoring for precision medicine [11-14], robotics and automation in home appliances [15], virtual online assistance [16], e”
Deep Learning for Spatio-Temporal Data Mining: A Survey (2019, background): “With the increasing availability of transportation data collected from various sensors like loop detector, road camera, and GPS, there is an urgent need to utilize deep learning methods to learn the complex and highly non-linear spatiotemporal correlations among the traffic data to facilitate variou”
T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction (2018, methodology): “Huang et al. [12] proposed a network architecture consisting of a deep belief network (DBN) and a regression model and veriﬁed that the network can capture random features from trafﬁc data on multiple datasets and this model improved prediction accuracy in trafﬁc forecasting.”
Short-Term Forecasting of Passenger Demand under On-Demand Ride Services: A Spatio-Temporal Deep Learning Approach (2017, background): “Huang et al. (2014) predicted short-term traffic flow via a two-layer DL structure with a deep belief network (DBN) at the bottom and a multitask regression model (MTL) at the top.”
Time Series Forecasting Based on Augmented Long Short-Term Memory (2017, neutral): “Deep belief networks (DBN) are frequently employed in current short-term traffic forecasting [7] [8], and pre-training strategies with unsupervised learning algorithms such as Restricted Boltzmann machine (RBM) [9] and Stacked AutoEncoder (SAE) [11] are also used.”
Rough Deep Neural Architecture for Short-Term Wind Speed Forecasting (2017, background): “One reason might be the fact that searching for optimal solution in the parameter space of ANNs with more than one hidden layer is a difficult task [17].”
Future Intelligent and Secure Vehicular Network Toward 6G: Machine-Learning Approaches (2020, methodology): “The spatial–temporal feature extraction-enabled ML, such as DBN, recurrent neural network (RNN), stacked autoencoders (SAEs), and CNN, is widely used in [88]–[90].”
Deep Multitask Learning for Railway Track Inspection (2015, methodology): “Material classification, segmentation and tie assessment algorithm is described in section V. Fastener detection and assessment algorithm is described in section VI. Experimental results are presented in section VII, and section VIII concludes the paper with a brief summary and discussion.”