Spatial Meta Learning With Comprehensive Prior Knowledge Injection for Service Time Prediction
Authors: 
Shuliang Wang, Qianyu Yang, Sijie Ruan, Cheng Long, Ye Yuan, Qi Li, Ziqiang Yuan, Jie Bao, 
Yu ZhengAuthors Info & Claims
Pages 936 - 950
Abstract
Intelligent logistics relies on accurately predicting the service time, which is a part of time cost in the last-mile delivery. However, service time prediction (STP) is non-trivial given complex delivery circumstances, location heterogeneity, and skewed observations in space, which are not well-handled by existing solutions. In our prior work, we treat STP at each location as a learning task to keep the location heterogeneity, propose a prior knowledge-enhanced meta-learning to tackle skewed observations, and introduce a Transformer-based representation module to encode complex delivery circumstances. Maintaining the design principles of prior work, in this extended paper, we propose MetaSTP<sup>+</sup>. In addition to fusing the prior knowledge after the meta-learning process, MetaSTP<sup>+</sup> also injects the prior knowledge before and during the meta-learning process to better tackle skewed observations. More specifically, MetaSTP<sup>+</sup> completes the support set of tasks with scarce samples from other tasks based on prior knowledge and is equipped with a prior knowledge-aware historical observation encoding module to achieve those purposes accordingly. Experiments show MetaSTP<sup>+</sup> outperforms the best baseline by 11.2% and 8.4% on two real-world datasets. Finally, an intelligent waybill assignment system based on MetaSTP<sup>+</sup> is deployed in JD Logistics.
References
[1]
M. Desrochers, J. Desrosiers, and M. Solomon, “A new optimization algorithm for the vehicle routing problem with time windows,” Operations Res., vol. 40, no. 2, pp. 342–354, 1992.
[2]
W. P. Nanry and J. Wesley Barnes, “Solving the pickup and delivery problem with time windows using reactive tabu search,” Transp. Res. Part B: Methodological, vol. 34, no. 2, pp. 107–121, 2000.
[3]
B. Guo et al., “Towards equitable assignment: Data-driven delivery zone partition at last-mile logistics,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2023, pp. 4078–4088.
[4]
F. Wu and L. Wu, “DeepETA: A spatial-temporal sequential neural network model for estimating time of arrival in package delivery system,” in Proc. AAAI Conf. Artif. Intell., 2019, pp. 774–781.
[5]
H. Zhang, H. Wu, W. Sun, and B. Zheng, “Deeptravel: A neural network based travel time estimation model with auxiliary supervision,” in Proc. Int. Joint Conf. Artif. Intell., 2018, pp. 3655–3661.
[6]
Z. Wang, K. Fu, and J. Ye, “Learning to estimate the travel time,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2018, pp. 858–866.
[7]
X. Fang, J. Huang, F. Wang, L. Liu, Y. Sun, and H. Wang, “SSML: Self-supervised meta-learner for en route travel time estimation at baidu maps,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2021, pp. 2840–2848.
[8]
S. Ruan et al., “Service time prediction for delivery tasks via spatial meta-learning,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2022, pp. 3829–3837.
[9]
S. Ruan et al., “Doing in one go: Delivery time inference based on couriers’ trajectories,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2020, pp. 2813–2821.
[10]
S. Ruan et al., “Discovering actual delivery locations from mis-annotated couriers’ trajectories,” in Proc. IEEE 38th Int. Conf. Data Eng., 2022, pp. 3241–3253.
[11]
J. Song, R. Wen, C. Xu, and J. W. E. Tay, “Service time prediction for last-yard delivery,” in Proc. IEEE Int. Conf. Big Data, 2019, pp. 3933–3938.
[12]
C. Finn, P. Abbeel, and S. Levine, “Model-agnostic meta-learning for fast adaptation of deep networks,” in Proc. Int. Conf. Mach. Learn., 2017, pp. 1126–1135.
[13]
T. Hospedales, A. Antoniou, P. Micaelli, and A. Storkey, “Meta-learning in neural networks: A survey,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 9, pp. 5149–5169, Sep. 2022.
[14]
Y. Liu, J. Ding, Y. Fu, and Y. Li, “UrbanKG: An urban knowledge graph system,” ACM Trans. Intell. Syst. Technol., vol. 14, no. 4, pp. 1–25, 2023.
[15]
Y. Liu, X. Zhang, J. Ding, Y. Xi, and Y. Li, “Knowledge-infused contrastive learning for urban imagery-based socioeconomic prediction,” in Proc. Int. Conf. World Wide Web, 2023, pp. 4150–4160.
[16]
Y. Yan et al., “UrbanCLIP: Learning text-enhanced urban region profiling with contrastive language-image pretraining from the web,” in Proc. Int. Conf. World Wide Web, 2024, pp. 4006–4017.
[17]
A. Vaswani et al., “Attention is all you need,” in Proc. Int. Conf. Neural Inf. Process. Syst., 2017, pp. 5998–6008.
[18]
N. Mishra, M. Rohaninejad, X. Chen, and P. Abbeel, “A simple neural attentive meta-learner,” in Proc. Int. Conf. Learn. Representations, 2018, pp. 1–17. [Online]. Available: https://openreview.net/forum?id=B1DmUzWAW
[19]
A. Van Den et al., “WaveNet: A generative model for raw audio,” Arxiv, pp. 1–15, 2016. [Online]. Available: https://research.google/pubs/wavenet-a-generative-model-for-raw-audio/
[20]
K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2016, pp. 770–778.
[21]
J. H. Friedman, “Greedy function approximation: A gradient boosting machine,” Ann. Statist., vol. 29, no. 5, pp. 1189–1232, 2001. [Online]. Available: https://www.jstor.org/stable/2699986
[22]
I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT, 2016.
[23]
H. Guo, R. Tang, Y. Ye, Z. Li, and X. He, “DeepFM: A factorization-machine based neural network for CTR prediction,” in Proc. Int. Joint Conf. Artif. Intell., 2017, pp. 1725–1731.
[24]
N. S. Altman, “An introduction to kernel and nearest-neighbor nonparametric regression,” Amer. Statistician, vol. 46, no. 3, pp. 175–185, 1992.
[25]
Y. Yuan, C. Shao, J. Ding, D. Jin, and Y. Li, “Spatio-temporal few-shot learning via diffusive neural network generation,” in Proc. Int. Conf. Learn. Representations, 2024, pp. 1–28. [Online]. Available: https://openreview.net/forum?id=QyFm3D3Tzi
[26]
A. G. Kek, R. L. Cheu, and Q. Meng, “Distance-constrained capacitated vehicle routing problems with flexible assignment of start and end depots,” Math. Comput. Modelling, vol. 47, no. 1/2, pp. 140–152, 2008.
[27]
J. Chen, Z. Xiao, D. Wang, W. Long, J. Bai, and V. Havyarimana, “Stay time prediction for individual stay behavior,” IEEE Access, vol. 7, pp. 130085–130100, 2019.
[28]
S. Liu, H. Cao, L. Li, and M. Zhou, “Predicting stay time of mobile users with contextual information,” IEEE Trans. Automat. Sci. Eng., vol. 10, no. 4, pp. 1026–1036, Oct. 2013.
[29]
G. Gidófalvi and F. Dong, “When and where next: Individual mobility prediction,” in Proc. 1st ACM SIGSPATIAL Int. Workshop Mobile Geographic Inf. Syst., 2012, pp. 57–64.
[30]
J. Han et al., “iETA: A robust and scalable incremental learning framework for time-of-arrival estimation,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2023, pp. 4100–4111.
[31]
H. Liu, W. Jiang, S. Liu, and X. Chen, “Uncertainty-aware probabilistic travel time prediction for on-demand ride-hailing at didi,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2023, pp. 4516–4526.
[32]
C. Gao et al., “A deep learning method for route and time prediction in food delivery service,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2021, pp. 2879–2889.
[33]
L. Zhang et al., “Delivery time prediction using large-scale graph structure learning based on quantile regression,” in Proc. IEEE Int. Conf. Data Eng., 2023, pp. 3403–3416.
[34]
Y. Qiang et al., “Modeling intra- and inter-community information for route and time prediction in last-mile delivery,” in Proc. Int. Conf. Data Eng., 2023, pp. 3106–3112.
[35]
H. Wen et al., “Enough waiting for the couriers: Learning to estimate package pick-up arrival time from couriers’ spatial-temporal behaviors,” ACM Trans. Intell. Syst. Technol., vol. 14, pp. 1–22, 2023.
[36]
T. Cai et al., “M2G4RTP: A multi-level and multi-task graph model for instant-logistics route and time joint prediction,” in Proc. Int. Conf. Data Eng., 2023, pp. 3296–3308.
[37]
H. Yao, Y. Liu, Y. Wei, X. Tang, and Z. Li, “Learning from multiple cities: A meta-learning approach for spatial-temporal prediction,” in Proc. Int. Conf. World Wide Web, 2019, pp. 2181–2191.
[38]
H. Qin, S. Ke, X. Yang, H. Xu, X. Zhan, and Y. Zheng, “Robust spatio-temporal purchase prediction via deep meta learning,” Proc. AAAI Conf. Artif. Intell., vol. 35, no. 5, 2021, pp. 4312–4319.
[39]
Y. Xu et al., “Metaptp: An adaptive meta-optimized model for personalized spatial trajectory prediction,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2022, pp. 2151–2159.
[40]
R. Jiang et al., “Spatio-temporal meta-graph learning for traffic forecasting,” in Proc. AAAI Conf. Artif. Intell., 2023, pp. 8078–8086.
[41]
M. Fan, Y. Sun, J. Huang, H. Wang, and Y. Li, “Meta-learned spatial-temporal poi auto-completion for the search engine at baidu maps,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2021, pp. 2822–2830.
[42]
Y. Chen, X. Wang, M. Fan, J. Huang, S. Yang, and W. Zhu, “Curriculum meta-learning for next POI recommendation,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2021, pp. 2692–2702.
[43]
X. Ding et al., “Delivery scope: A new way of restaurant retrieval for on-demand food delivery service,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2020, pp. 3026–3034.
[44]
H. Wen et al., “Package pick-up route prediction via modeling couriers’ spatial-temporal behaviors,” in Proc. IEEE Int. Conf. Data Eng., 2021, pp. 2141–2146.
[45]
H. Wen et al., “DeepRoute : Modeling couriers’ spatial-temporal behaviors and decision preferences for package pick-up route prediction,” ACM Trans. Intell. Syst. Technol., vol. 13, no. 2, pp. 1–23, 2022.
[46]
H. Wen et al., “Graph2Route: A dynamic spatial-temporal graph neural network for pick-up and delivery route prediction,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2022, pp. 4143–4152.
[47]
M. Dahiya, D. Samatia, and K. Rustogi, “Learning locality maps from noisy geospatial labels,” in Proc. 35th Annu. ACM Symp. Appl. Comput., 2020, pp. 601–608.
[48]
V. Srivastava, P. Tejaswin, L. Dhakad, M. Kumar, and A. Dani, “A geocoding framework powered by delivery data,” in Proc. 28th Int. Conf. Adv. Geographic Inf. Syst., 2020, pp. 568–577.
[49]
Y. Song, J. Li, L. Chen, S. Chen, R. He, and Z. Sun, “A semantic segmentation based POI coordinates generating framework for on-demand food delivery service,” in Proc. 29th Int. Conf. Adv. Geographic Inf. Syst., 2021, pp. 379–388.
[50]
D. Jiang et al., “ALWAES: An automatic outdoor location-aware correction system for online delivery platforms,” Proc. ACM Interactive Mobile Wearable Ubiquitous Technol., vol. 5, no. 3, pp. 1–24, 2021.
[51]
Y. Ding, D. Jiang, Y. Yang, Y. Liu, T. He, and D. Zhang, “P2-loc: A person-2-person indoor localization system in on-demand delivery,” ACM Interactive, Mobile, Wearable Ubiquitous Technol., vol. 6, pp. 1–24, 2022.
Index Terms
- Spatial Meta Learning With Comprehensive Prior Knowledge Injection for Service Time Prediction
Index terms have been assigned to the content through auto-classification.
Recommendations
Risk Prediction on Electronic Health Records with Prior Medical Knowledge
KDD '18: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data MiningPredicting the risk of potential diseases from Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Compared with traditional machine learning models, deep ...
Service Time Prediction for Delivery Tasks via Spatial Meta-Learning
KDD '22: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data MiningService time is a part of time cost in the last-mile delivery, which is the time spent on delivering parcels at a certain location. Predicting the service time is fundamental for many downstream logistics applications, e.g., route planning with time ...
Urban Traffic Dynamics Prediction—A Continuous Spatial-temporal Meta-learning Approach
Urban traffic status (e.g., traffic speed and volume) is highly dynamic in nature, namely, varying across space and evolving over time. Thus, predicting such traffic dynamics is of great importance to urban development and transportation management. ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
1041-4347 © 2024 IEEE. All rights reserved, including rights for text and data mining, and training of artificial intelligence and similar technologies. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.
Publisher
IEEE Educational Activities Department
United States
Publication History
Published: 01 February 2025
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 17 Feb 2025