Diffusion Convolutional Recurrent Neural Network

Diffusion Convolutional Recurrent Neural Networks (DCRNNs) are a class of deep learning models designed to analyze spatiotemporal data by combining the strengths of recurrent neural networks (for temporal dependencies) and graph convolutional networks (for spatial relationships). Current research focuses on improving DCRNN performance through techniques like self-supervised pretraining and deep ensemble methods for uncertainty quantification, as well as exploring alternative architectures such as Graph Convolutional Gated Recurrent Networks (GCGRNNs) for enhanced efficiency and accuracy. These models find applications in diverse fields, including traffic forecasting, seizure detection from EEG data, and infectious disease modeling, offering improved prediction accuracy and uncertainty estimation compared to previous methods.

Papers

November 5, 2024

Pedestrian Volume Prediction Using a Diffusion Convolutional Gated Recurrent Unit Model
Yiwei Dong, Tingjin Chu, Lele Zhang, Hadi Ghaderi, Hanfang Yang
Gated Recurrent Unit Recurrent Unit Pedestrian Crossing Prediction Crowd Flow Pedestrian Prediction Model Diffusion Convolutional Recurrent Neural Network

October 28, 2024

Graph Based Traffic Analysis and Delay Prediction
Gabriele Borg, Charlie Abela
Graph Drawing Traffic Data Traffic Datasets Traffic Efficiency Delay Prediction Diffusion Convolutional Recurrent Neural Network

November 7, 2023

RobustMat: Neural Diffusion for Street Landmark Patch Matching under Challenging Environments
Rui She, Qiyu Kang, Sijie Wang, Yuan-Rui Yang, Kai Zhao, Yang Song, Wee Peng Tay
Autonomous Vehicle Challenging Environment Neural Diffusion Diffusion Convolutional Recurrent Neural Network

June 28, 2022

Improving self-supervised pretraining models for epileptic seizure detection from EEG data
Sudip Das, Pankaj Pandey, Krishna Prasad Miyapuram
Self Supervision EEG Data Temple University Hospital Diffusion Convolutional Recurrent Neural Network

April 4, 2022

Deep-Ensemble-Based Uncertainty Quantification in Spatiotemporal Graph Neural Networks for Traffic Forecasting
Tanwi Mallick, Prasanna Balaprakash, Jane Macfarlane
Uncertainty Quantification Traffic Forecasting Convolutional Recurrent Short Term Traffic Spatiotemporal Graph Neural Network Diffusion Convolutional Recurrent Neural Network

November 22, 2021

Network-wide Multi-step Traffic Volume Prediction using Graph Convolutional Gated Recurrent Neural Network
Lei Lin, Weizi Li, Lei Zhu
Graph Neural Network Traffic Data Network Wide Traffic Diffusion Convolutional Recurrent Neural Network

November 9, 2021

Deep diffusion-based forecasting of COVID-19 by incorporating network-level mobility information
Padmaksha Roy, Shailik Sarkar, Subhodip Biswas, Fanglan Chen, Zhiqian Chen, Naren Ramakrishnan, Chang-Tien Lu
Diffusion Model Time Series Covid 19 Diffusion Convolutional Recurrent Neural Network

Diffusion Convolutional Recurrent Neural Network

Papers

Pedestrian Volume Prediction Using a Diffusion Convolutional Gated Recurrent Unit Model

Graph Based Traffic Analysis and Delay Prediction

RobustMat: Neural Diffusion for Street Landmark Patch Matching under Challenging Environments

Improving self-supervised pretraining models for epileptic seizure detection from EEG data

Deep-Ensemble-Based Uncertainty Quantification in Spatiotemporal Graph Neural Networks for Traffic Forecasting

Network-wide Multi-step Traffic Volume Prediction using Graph Convolutional Gated Recurrent Neural Network

Deep diffusion-based forecasting of COVID-19 by incorporating network-level mobility information