More Web Proxy on the site http://driver.im/

research-article

Open access

MM-DAG: Multi-task DAG Learning for Multi-modal Data - with Application for Traffic Congestion Analysis

Authors:

Wolfgang Ketter,

Chen ZhangAuthors Info & Claims

KDD '23: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

Pages 1188 - 1199

https://doi.org/10.1145/3580305.3599436

Published: 04 August 2023 Publication History

Abstract

This paper proposes to learn Multi-task, Multi-modal Direct Acyclic Graphs (MM-DAGs), which are commonly observed in complex systems, e.g., traffic, manufacturing, and weather systems, whose variables are multi-modal with scalars, vectors, and functions. This paper takes the traffic congestion analysis as a concrete case, where a traffic intersection is usually regarded as a DAG. In a road network of multiple intersections, different intersections can only have someoverlapping and distinct variables observed. For example, a signalized intersection has traffic light-related variables, whereas unsignalized ones do not. This encourages the multi-task design: with each DAG as a task, the MM-DAG tries to learn the multiple DAGs jointly so that their consensus and consistency are maximized. To this end, we innovatively propose a multi-modal regression for linear causal relationship description of different variables. Then we develop a novel Causality Difference (CD) measure and its differentiable approximator. Compared with existing SOTA measures, CD can penalize the causal structural difference among DAGs with distinct nodes and can better consider the uncertainty of causal orders. We rigidly prove our design's topological interpretation and consistency properties. We conduct thorough simulations and one case study to show the effectiveness of our MM-DAG. The code is available under https://github.com/Lantian72/MM-DAG.

Supplementary Material

MP4 File (e6c1154f5b7cae8728cab7c53183b270.mp4)

Presentation video

Download
3.53 MB

MP4 File (20230626_205728.mp4)

promotional video

Download
3.78 MB

References

[1]

Tanzina Afrin and Nita Yodo. 2021. A probabilistic estimation of traffic congestion using Bayesian network. Measurement 174 (2021), 109051.

[2]

Ranwa Al Mallah, Alejandro Quintero, and Bilal Farooq. 2017. Distributed clas-sification of urban congestion using VANET. IEEE Transactions on Intelligent Transportation Systems 18, 9 (2017), 2435--2442.

Digital Library

[3]

Ranwa Al Mallah, Alejandro Quintero, and Bilal Farooq. 2019. Cooperative evaluation of the cause of urban traffic congestion via connected vehicles. IEEE Transactions on Intelligent Transportation Systems 21, 1 (2019), 59--67.

[4]

Michael Behrisch, Laura Bieker, Jakob Erdmann, and Daniel Krajzewicz. 2011. SUMO--simulation of urban mobility: an overview. In Proceedings of SIMUL 2011, The Third International Conference on Advances in System Simulation. ThinkMind.

[5]

Rohit Bhattacharya, Tushar Nagarajan, Daniel Malinsky, and Ilya Shpitser. 2021. Differentiable causal discovery under unmeasured confounding. In International Conference on Artificial Intelligence and Statistics. PMLR, 2314--2322.

[6]

Xinshi Chen, Haoran Sun, Caleb Ellington, Eric Xing, and Le Song. 2021. Multi-task Learning of Order-Consistent Causal Graphs. Advances in Neural Information Processing Systems 34 (2021), 11083--11095.

[7]

Yirong Chen, Ziyue Li, Wanli Ouyang, and Michael Lepech. 2023. Adaptive Hierarchical SpatioTemporal Network for Traffic Forecasting. arXiv preprint arXiv:2306.09386 (2023).

[8]

David Maxwell Chickering. 2002. Optimal structure identification with greedy search. Journal of machine learning research 3, Nov (2002), 507--554.

[9]

Andy HF Chow, Alex Santacreu, Ioannis Tsapakis, Garavig Tanasaranond, and Tao Cheng. 2014. Empirical assessment of urban traffic congestion. Journal of advanced transportation 48, 8 (2014), 1000--1016.

[10]

Paul Erdos, Alfréd Rényi, et al. 1960. On the evolution of random graphs. Publ. Math. Inst. Hung. Acad. Sci 5, 1 (1960), 17--60.

[11]

Xinyue Fan, Jiao Zhang, and Qi Shen. 2019. Prediction of road congestion diffusion based on dynamic Bayesian networks. In Journal of Physics: Conference Series, Vol. 1176. IOP Publishing, 022046.

[12]

Steven Finch. 2003. Transitive relations, topologies and partial orders. unpublished note (2003).

[13]

Erdun Gao, Junjia Chen, Li Shen, Tongliang Liu, Mingming Gong, and Howard Bondell. 2021. FedDAG: Federated DAG Structure Learning. Transactions on Machine Learning Research (2021).

[14]

Naftali Harris and Mathias Drton. 2013. PC algorithm for nonparanormal graphical models. Journal of Machine Learning Research 14, 11 (2013).

[15]

John Gilbert Hocking and Gail S Young. 1988. Topology. Courier Corporation.

[16]

Min Jiang, Andi Wang, Ziyue Li, and Fugee Tsung. 2023. A Unified Probabilistic Framework for Spatiotemporal Passenger Crowdedness Inference within Urban Rail Transit Network. arXiv preprint arXiv:2306.08343 (2023).

[17]

Yunlong Jiao and Jean-Philippe Vert. 2015. The Kendall and Mallows kernels for permutations. In International Conference on Machine Learning. PMLR, 1935--1944.

[18]

Jiwon Kim and Guangxing Wang. 2016. Diagnosis and prediction of traffic congestion on urban road networks using Bayesian networks. Transportation Research Record 2595, 1 (2016), 108--118.

[19]

Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014).

[20]

Mikko Koivisto and Kismat Sood. 2004. Exact Bayesian structure discovery in Bayesian networks. The Journal of Machine Learning Research 5 (2004), 549--573.

Digital Library

[21]

Ziyue Li, Nurettin Dorukhan Sergin, Hao Yan, Chen Zhang, and Fugee Tsung. 2020. Tensor completion for weakly-dependent data on graph for metro passenger flow prediction. In proceedings of the AAAI conference on artificial intelligence, Vol. 34. 4804--4810.

[22]

Ziyue Li, Hao Yan, Fugee Tsung, and Ke Zhang. 2022. Profile Decomposition Based Hybrid Transfer Learning for Cold-Start Data Anomaly Detection. ACM Transactions on Knowledge Discovery from Data (TKDD) 16, 6 (2022), 1--28.

[23]

Ziyue Li, Hao Yan, Chen Zhang, and Fugee Tsung. 2020. Long-short term spatiotemporal tensor prediction for passenger flow profile. IEEE Robotics and Automation Letters 5, 4 (2020), 5010--5017.

[24]

Ziyue Li, Hao Yan, Chen Zhang, and Fugee Tsung. 2022. Individualized passenger travel pattern multi-clustering based on graph regularized tensor latent dirichlet allocation. Data Mining and Knowledge Discovery 36, 4 (2022), 1247--1278.

Digital Library

[25]

Junpeng Lin, Ziyue Li, Zhishuai Li, Lei Bai, Rui Zhao, and Chen Zhang. 2023. Dynamic Causal Graph Convolutional Network for Traffic Prediction. arXiv preprint arXiv:2306.07019 (2023).

[26]

Jiancheng Long, Ziyou Gao, Xiaomei Zhao, Aiping Lian, and Penina Orenstein. 2011. Urban traffic jam simulation based on the cell transmission model. Networks and Spatial Economics 11, 1 (2011), 43--64.

[27]

Sen Luan, Ruimin Ke, Zhou Huang, and Xiaolei Ma. 2022. Traffic congestion prop-agation inference using dynamic Bayesian graph convolution network. Transportation research part C: emerging technologies 135 (2022), 103526.

[28]

Zhenyu Mao, Ziyue Li, Dedong Li, Lei Bai, and Rui Zhao. 2022. Jointly Contrastive Representation Learning on Road Network and Trajectory. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management. 1501--1510.

Digital Library

[29]

Preetam Nandy, Alain Hauser, and Marloes H Maathuis. 2018. High-dimensional consistency in score-based and hybrid structure learning. The Annals of Statistics 46, 6A (2018), 3151--3183.

[30]

Ignavier Ng, AmirEmad Ghassami, and Kun Zhang. 2020. On the role of sparsity and dag constraints for learning linear dags. Advances in Neural Information Processing Systems 33 (2020), 17943--17954.

[31]

Ignavier Ng, Sébastien Lachapelle, Nan Rosemary Ke, Simon Lacoste-Julien, and Kun Zhang. 2022. On the convergence of continuous constrained optimization for structure learning. In International Conference on Artificial Intelligence and Statistics. PMLR, 8176--8198.

[32]

Alexandru Niculescu-Mizil and Rich Caruana. 2007. Inductive transfer for Bayesian network structure learning. In Artificial intelligence and statistics. PMLR, 339--346.

[33]

Chris J Oates, Jim Q Smith, Sach Mukherjee, and James Cussens. 2016. Exact estimation of multiple directed acyclic graphs. Statistics and Computing 26, 4 (2016), 797--811.

Digital Library

[34]

Diane Oyen and Terran Lane. 2012. Leveraging domain knowledge in multitask Bayesian network structure learning. In Twenty-Sixth AAAI conference on artificial intelligence.

[35]

Diane Oyen and Terran Lane. 2013. Bayesian discovery of multiple Bayesian networks via transfer learning. In 2013 IEEE 13th International Conference on Data Mining. IEEE, 577--586.

[36]

Sebastian Raschka. 2014. An overview of general performance metrics of binary classifier systems. arXiv preprint arXiv:1410.5330 (2014).

[37]

Gonzalo A Ruz, Pablo A Henríquez, and Aldo Mascareño. 2020. Sentiment analysis of Twitter data during critical events through Bayesian networks classifiers. Future Generation Computer Systems 106 (2020), 92--104.

Digital Library

[38]

Bo Shen, Raghav Gnanasambandam, Rongxuan Wang, and Zhenyu James Kong. 2023. Multi-task Gaussian process upper confidence bound for hyperparameter tuning and its application for simulation studies of additive manufacturing. IISE Transactions 55, 5 (2023), 496--508.

[39]

Peter Spirtes, Clark N Glymour, Richard Scheines, and David Heckerman. 2000. Causation, prediction, and search. MIT press.

[40]

Shiliang Sun, Changshui Zhang, and Guoqiang Yu. 2006. A Bayesian network approach to traffic flow forecasting. IEEE Transactions on intelligent transportation systems 7, 1 (2006), 124--132.

Digital Library

[41]

Marina Velikova, Josien Terwisscha van Scheltinga, Peter JF Lucas, and Marc Spaanderman. 2014. Exploiting causal functional relationships in Bayesian network modelling for personalised healthcare. International Journal of Approximate Reasoning 55, 1 (2014), 59--73.

[42]

Luxuan Wang, Lei Bai, Ziyue Li, Rui Zhao, and Fugee Tsung. 2023. Correlated Time Series Self-Supervised Representation Learning via Spatiotemporal Bootstrapping. arXiv preprint arXiv:2306.06994 (2023).

[43]

Peter Wills and François G Meyer. 2020. Metrics for graph comparison: a practi-tioner's guide. Plos one 15, 2 (2020), e0228728.

[44]

Lei Xing, Wenjun Wang, Guixiang Xue, Hao Yu, Xiaotong Chi, and Weidi Dai. 2015. Discovering traffic outlier causal relationship based on anomalous DAG. In International Conference in Swarm Intelligence. Springer, 71--80.

[45]

Fang Yao, Hans-Georg Müller, and Jane-Ling Wang. 2005. Functional linear regression analysis for longitudinal data. The Annals of Statistics 33, 6 (2005), 2873--2903.

[46]

Junwen Yao, Jonas Mueller, and Jane-Ling Wang. 2021. Deep learning for functional data analysis with adaptive basis layers. In International Conference on Machine Learning. PMLR, 11898--11908.

[47]

Yue Yu, Jie Chen, Tian Gao, and Mo Yu. 2019. DAG-GNN: DAG structure learning with graph neural networks. In International Conference on Machine Learning. PMLR, 7154--7163.

[48]

Aomuhan Zhang and Ziyou Gao. 2012. CTM-based Propagation of Non-recurrent Congestion and Location of Variable Message Sign. In 2012 Fifth International Joint Conference on Computational Sciences and Optimization. IEEE, 462--465.

[49]

Yu Zhang and Qiang Yang. 2021. A survey on multi-task learning. IEEE Transactions on Knowledge and Data Engineering 34, 12 (2021), 5586--5609.

[50]

Xun Zheng, Bryon Aragam, Pradeep K Ravikumar, and Eric P Xing. 2018. Dags with no tears: Continuous optimization for structure learning. Advances in Neural Information Processing Systems 31 (2018).

[51]

Xun Zheng, Chen Dan, Bryon Aragam, Pradeep Ravikumar, and Eric Xing. 2020. Learning sparse nonparametric dags. In International Conference on Artificial Intelligence and Statistics. PMLR, 3414--3425.

[52]

Yun Zhou, Jiang Wang, Cheng Zhu, and Weiming Zhang. 2017. Multiple dags learning with non-negative matrix factorization. In Advanced Methodologies for Bayesian Networks. PMLR, 81--92.

[53]

LI Ziyue. 2021. Tensor topic models with graphs and applications on individual-ized travel patterns. In 2021 IEEE 37th International Conference on Data Engineering (ICDE). IEEE, 2756--2761.

Cited By

Wang LBai LLi ZZhao RTsung F(2023)Correlated Time Series Self-Supervised Representation Learning via Spatiotemporal Bootstrapping2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)10.1109/CASE56687.2023.10260640(1-7)Online publication date: 26-Aug-2023
https://doi.org/10.1109/CASE56687.2023.10260640
Lin JLi ZLi ZBai LZhao RZhang C(2023)Dynamic Causal Graph Convolutional Network for Traffic Prediction2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)10.1109/CASE56687.2023.10260564(1-8)Online publication date: 26-Aug-2023
https://doi.org/10.1109/CASE56687.2023.10260564

Index Terms

MM-DAG: Multi-task DAG Learning for Multi-modal Data - with Application for Traffic Congestion Analysis
1. Computing methodologies
  1. Machine learning
    1. Learning paradigms
      1. Multi-task learning
2. Mathematics of computing
  1. Probability and statistics
    1. Probabilistic representations
      1. Causal networks

Recommendations

A New Skeleton-Neural DAG Learning Approach
Advances in Knowledge Discovery and Data Mining
Abstract
Learning a Directed Acyclic Graph (DAG) structure from observational data plays an essential role in causal inference and machine learning. A recent advance in the area is that the DAG learning problem was formulated as a continuous optimization ...
Beware of the DAG!
COA'08: Proceedings of the 2008th International Conference on Causality: Objectives and Assessment - Volume 6

Directed acyclic graph (DAG) models are popular tools for describing causal relationships and for guiding attempts to learn them from data. They appear to supply a means of extracting causal conclusions from probabilistic conditional independence ...
Learning quadratic variance function (QVF) DAG models via overdispersion scoring (ODS)

Learning DAG or Bayesian network models is an important problem in multi-variate causal inference. However, a number of challenges arises in learning large-scale DAG models including model identifiability and computational complexity since the space of ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image ACM Conferences

KDD '23: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 2023

5996 pages

ISBN:9798400701030

DOI:10.1145/3580305

General Chairs:
Ambuj Singh
UC Santa Barbara, USA
,
Yizhou Sun
UC Los Angeles, USA
,
Program Chairs:
Leman Akoglu
Carnegie Mellon University, USA
,
Dimitrios Gunopulos
University of Athens, Greece
,
Xifeng Yan
UC Santa Barbara, USA
,
Ravi Kumar
Google, USA
,
Fatma Ozcan
Google, USA
,
Jieping Ye
Alibaba DAMO Academy

Copyright © 2023 Owner/Author.

This work is licensed under a Creative Commons Attribution International 4.0 License.

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 04 August 2023

Check for updates

Author Tags

Qualifiers

Research-article

Funding Sources

BNSF
NSFC

Conference

KDD '23

Sponsor:

KDD '23: The 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 6 - 10, 2023

CA, Long Beach, USA

Acceptance Rates

Overall Acceptance Rate 1,133 of 8,635 submissions, 13%

Upcoming Conference

KDD '25

Sponsor:
sigkdd
sigkdd

The 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 3 - 7, 2025

Toronto , ON , Canada

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

2
Total Citations
View Citations
801
Total Downloads

Downloads (Last 12 months)618
Downloads (Last 6 weeks)66

Reflects downloads up to 19 Dec 2024

Other Metrics

View Author Metrics

Citations

Cited By

Wang LBai LLi ZZhao RTsung F(2023)Correlated Time Series Self-Supervised Representation Learning via Spatiotemporal Bootstrapping2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)10.1109/CASE56687.2023.10260640(1-7)Online publication date: 26-Aug-2023
https://doi.org/10.1109/CASE56687.2023.10260640
Lin JLi ZLi ZBai LZhao RZhang C(2023)Dynamic Causal Graph Convolutional Network for Traffic Prediction2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)10.1109/CASE56687.2023.10260564(1-8)Online publication date: 26-Aug-2023
https://doi.org/10.1109/CASE56687.2023.10260564

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Media

Figures

Other

Tables

View Table of Contents