[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traffic Congestion Event Prediction.pptx

Spatio-Temporal Graph Neural Point Process for Traffic Congestion Event Prediction

1. Quang-Huy Tran Network Science Lab Dept. of Artificial Intelligence The Catholic University of Korea E-mail: [email protected] 2024-04-29 Spatio -Temporal Graph Neural Point Process for Traffic Congestion Event Prediction Guangyin Jin et al. AAAI’37: 2023 Conference on Artificial Intelligence

2. OUTLINE MOTIVATION INTRODUCTION PROBLEM FORMULATION METHODOLOGY EXPERIMENT & RESULT CONCLUSION

3. MOTIVATION Traffic congestion is one of the most serious problems in urban management . Traffic congestion is a continuous process from generation to dissipation. I ndividual congestion event: occurrence time and duration . Meaningful for prediction to improve the traffic management and scheduling . when the next congestion event occur. how long it will last. Traffic congestion overview Previous have disadvantages: The conventional methods only model dense variables like road, sparse like congestion not done. support the prediction in the given future time window (short time), not suitable for congestion (long time) .

4. MOTIVATION An appropriate framework for sparse event prediction in continuous-time . Neural Point Process Challenges: 1) How to effectively capture the spatio -temporal dependencies inroad networks?. 2) How to effectively model the continuous and instantaneous temporal dynamics simultaneously for each road? Probabilistic models of variable-length point sequences observed on the real half-line—here interpreted as arrival times of events .

5. INTRODUCTION Propose a novel model named Spatio -Temporal Graph Neural Point Process (STGNPP) for traffic congestion event prediction. Transformer and Graph Convolution Network (GCN) to jointly capture the spatio -temporal dependencies from traffic states data. Extract the contextual link representations to incorporate with congestion event information for modeling the history of the point process. To encode the hidden evolution patterns of each road present a novel continuous Gated Recurrent Unit (GRU) layer with neural flow architecture. First work to propose spatio -temporal graph neural point process.

6. METHODOLOGY Task definition A road network with links as a graph   Traffic states ( eg. , link speed) on each link are dense features in the snapshots of certain time granularity.   Given a fixed-length historical time window T for each sample : predict the occurrence time and duration of the next congestion event . Sequential congestion events : Link has . occurrence time . : duration .  

7. METHODOLOGY Point Process Distribution Stochastic process to simulate the sequential events in a given observation time interval   Time point is given: Intensity function of events at time point depended on the historical sequential events up to :   Probability density function to observe an event sequence , inter-event time:  

8. METHODOLOGY Overall Architecture

9. METHODOLOGY Spatio -Temporal Graph Learning Module Link-wise Transformer layer. Graph convolution layer. Spatio -temporal inquirer. First, a fully connected layer to map the historical traffic states into high-dimensional representation.

10. METHODOLOGY Link-Wise Transformer Layer Self-attention network: Employ trigonometric functions-based position encoding method. where and are query, key, and value matrices obtained by three linear transformations are dimension.   Pass into two-layer position-wise feed-forward neural network. mask operation that sets the value of the upper triangle of attention matrix to 0  

11. METHODOLOGY Graph Convolution Layer - Spatio -temporal Inquirer Simple graph convolution operation with mix-hop aggregation. where A is the normalized predefined adjacency matrix, are two learnable matrices, is learnable weight for each convolution layer.   Select corresponding hidden representations based on indexes. Obtain those representation using sum aggregation and zero padding.

12. METHODOLOGY Congestion Event Learning Module – Continuous GRU Layer Congestion event representation where denotes the historical duration of each congestion event after zero padding. Insight: the traffic state for each link is a combination of continuous changes and instantaneous change. Apply ODE-based( ordinary differential equations ) : where is a continuous function satisfies 2 properties: i ) and ii) , is an arbitrary contractive neural network.  

13. METHODOLOGY Congestion Event Learning Module – Continuous GRU Layer Apply GRU-ODE Continuous GRU. Instantaneous dynamics GRU.

14. METHODOLOGY Optimization and Prediction Optimizes the negative log-likelihood of the probability density function of the inter-event time and the absolute error of the duration prediction: where denotes the fully connected layer for duration prediction of the next traffic congestion, denotes the tradeoff ratio.   Intensity Function Network: To approximate the distribution of inter-event time and characterize the effect of periodic patterns of congestion, a periodic gated unit to adjust the intensity function is defined:

15. EXPERIMENT AND RESULT EXPERIMENT Measurement: Mean Absolute Errors (MAE). Negative log-likelihood (NLL). Dataset: Amap application Beijing and Chengdu . interevent times, duration and periodic features. Task: Predict link condition in next 6 hours.

16. Baseline: Simple model: Historical Average (HA), Gradient Boosting Decision Tree(GBDT ) [1] , Gate Recurrent Unit (GRU) [2] . Spatio -temporal GNN: DCRNN [3] , GraphWaveNet [4] , STGODE [5] . Neural point-process model: NHTPP [6] , RMTPP [7] , THPP [8] , FNN-TPP [9] . EXPERIMENT AND RESULT EXPERIMENT [1] Ye, J., Chow, J. H., Chen, J., & Zheng, Z. (2009, November). Stochastic gradient boosted distributed decision trees. In Proceedings of the 18th ACM conference on Information and knowledge management (pp. 2061-2064).. [2] Cho, K., Van Merriënboer , B., Gulcehre , C., Bahdanau , D., Bougares , F., Schwenk , H., & Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078. [3] Li, Y., Yu, R., Shahabi , C., & Liu, Y. (2017). Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. arXiv preprint arXiv:1707.01926. [4] Wu, Z., Pan, S., Long, G., Jiang, J., & Zhang, C. (2019). Graph wavenet for deep spatial-temporal graph modeling. arXiv preprint arXiv:1906.00121. [5] Fang, Z., Long, Q., Song, G., & Xie, K. (2021, August). Spatial-temporal graph ode networks for traffic flow forecasting. In Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining (pp. 364-373). [6] Mei, H., & Eisner, J. M. (2017). The neural hawkes process: A neurally self-modulating multivariate point process. Advances in neural information processing systems, 30. [7] Du, N., Dai, H., Trivedi, R., Upadhyay, U., Gomez-Rodriguez, M., & Song, L. (2016, August). Recurrent marked temporal point processes: Embedding event history to vector. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1555-1564). [8] Zuo , S., Jiang, H., Li, Z., Zhao, T., & Zha, H. (2020, November). Transformer hawkes process. In International conference on machine learning (pp. 11692-11702). PMLR. [9] Omi, T., & Aihara , K. (2019). Fully neural network based model for general temporal point processes. Advances in neural information processing systems, 32.

17. EXPERIMENT AND RESULT RESULT – Overall Perfor mance

18. EXPERIMENT AND RESULT RESULT – Ablation study and Parameter study

19. CONCLUSION Propose a novel spatio -temporal graph neural point process framework for traffic congestion event prediction . utilize the spatiotemporal graph to incorporate with neural point process for traffic congestion event modeling. consider periodic features, continuous and instantaneous dynamics to improve the inter-event dependencies learning. Experiment shows that the proposed demonstrate the superiority compared with other traditional methods.