skip to main content
10.1145/3678717.3691246acmconferencesArticle/Chapter ViewAbstractPublication PagesgisConference Proceedingsconference-collections
short-paper
Open access

SRL: Towards a General-Purpose Framework for Spatial Representation Learning

Published: 22 November 2024 Publication History

Abstract

Representation learning (RL) techniques are widely adopted in areas such as natural language processing and computer vision, with prominent examples such as attention and ConvNet architectures. In comparison, many GeoAI works still rely on feature engineering or data conversion to represent spatial data (e.g., points, polylines, polygons, 3D building models, etc.) as features in formats that are easier for neural networks to handle. The neural network architectures remain unchanged, and the need for feature engineering has become a bottleneck for applying deep learning to new tasks in the age of big data. In this paper, we advocate the idea of developing learnable spatial representation modules, which not only enable spatial reasoning but also enable neural nets to directly consume (i.e., encoding) or generate (i.e., decoding) spatial data. We propose Spatial Representation Learning (SRL), a new general-purpose representation learning framework for spatial reasoning. We discuss the key challenges of spatial representation learning including multi-scale RL, continuous RL, shape-centric RL, noise-robust RL, heterogeneity-aware RL, and fairness-aware RL. We also discuss the critical role and potential of SRL in various geospatial subdomains and how this technique can lead to a new generation of GeoAI.

References

[1]
Andrés et al. 2013. Geo-indistinguishability: Differential privacy for location-based systems. In ACM SIGSAC 2013. 901--914.
[2]
Cheng Chi et al. 2023. Diffusion policy: Visuomotor policy learning via action diffusion. arXiv preprint arXiv:2303.04137 (2023).
[3]
Yezhen Cong et al. 2022. SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery. In NeurIPS 2022.
[4]
Ruiqi Gao et al. 2022. Learning V1 Simple Cells with Vector Representation of Local Content and Matrix Representation of Local Motion. In AAAI 2022.
[5]
Michael F Goodchild and Wenwen Li. 2021. Replication across space and time must be weak in the social and environmental sciences. PNAS 118, 35 (2021).
[6]
Erhu He et al. 2023. Physics Guided Neural Networks for Time-aware Fairness: An Application in Crop Yield Prediction. In AAAI 2023.
[7]
Erhu He et al. 2024. Learning With Location-Based Fairness: A Statistically-Robust Framework and Acceleration. TKDE (2024).
[8]
Christopher Irrgang et al. 2021. Towards neural Earth system modelling by integrating artificial intelligence in Earth system science. Nature Machine Intelligence 3, 8 (2021), 667--674.
[9]
Yuhao Kang and et al. 2019. Transferring multiscale map styles using generative adversarial networks. International Journal of Cartography 5, 2--3 (2019), 115--141.
[10]
Jina Kim et al. 2023. The mapKurator System: A Complete Pipeline for Extracting and Linking Text from Historical Maps. In ACM SIGSPATIAL 2023. 35.
[11]
Yochanan Kushnir et al. 2019. Towards operational predictions of the near-term climate. Nature Climate Change 9, 2 (2019), 94--101.
[12]
Hao Li et al. 2023. Rethink Geographical Generalizability with Unsupervised Self-Attention Model Ensemble: A Case Study of OpenStreetMap Missing Building Detection in Africa. In ACM SIGSPATIAL 2023.
[13]
Jiang Liu et al. 2023. Polyformer: Referring image segmentation as sequential polygon generation. In CVPR 2023. 18653--18663.
[14]
Oisin Mac Aodha et al. 2019. Presence-only geographical priors for fine-grained image classification. In ICCV 2019. 9596--9606.
[15]
Gengchen Mai et al. 2020. Multi-Scale Representation Learning for Spatial Feature Distributions using Grid Cells. In ICLR 2020. openreview.
[16]
Gengchen Mai et al. 2022. A review of location encoding for GeoAI: methods and applications. IJGIS 36, 4 (2022), 639--673.
[17]
Gengchen Mai et al. 2023. Csp: Self-supervised contrastive spatial pre-training for geospatial-visual representations. In ICML 2023. PMLR, 23498--23515.
[18]
Gengchen Mai et al. 2023. Towards general-purpose representation learning of polygonal geometries. GeoInformatica 27, 2 (2023), 289--340.
[19]
Jinmeng Rao et al. 2020. LSTM-TrajGAN: A Deep Learning Approach to Trajectory Privacy Protection. In GIScience 2020. 12:1--12:17.
[20]
Jinmeng Rao et al. 2021. A privacy-preserving framework for location recommendation using decentralized collaborative machine learning. TGIS (2021).
[21]
Jinmeng Rao et al. 2023. CATS: Conditional Adversarial Trajectory Synthesis for privacy-preserving trajectory data publication using deep learning approaches. International Journal of Geographical Information Science 37, 12 (2023), 2538--2574.
[22]
Kevin Tang et al. 2015. Improving image classification with location context. In ICCV 2015. 1008--1016.
[23]
Nam Vo et al. 2017. Revisiting im2gps in the deep learning era. In ICCV 2017.
[24]
Nemin Wu et al. 2024. TorchSpatial: A Location Encoding Framework and Benchmark for Spatial Representation Learning. arXiv preprint arXiv:2406.15658 (2024).
[25]
Yiqun Xie et al. 2021. A statistically-guided deep network transformation and moderation framework for data with spatial heterogeneity. In ICDM. 767--776.
[26]
Yiqun Xie et al. 2023. Harnessing Heterogeneity in Space with Statistically-Guided Meta-Learning. KAIS 65 (2023), 2699--2729.
[27]
Xiongfeng Yan et al. 2022. A graph deep learning approach for urban building grouping. Geocarto International 37, 10 (2022), 2944--2966.
[28]
Yuanwen Yue et al. 2023. Connecting the dots: Floorplan reconstruction using two-level queries. In CVPR 2023. 845--854.
[29]
Keer Zhang et al. 2023. A global dataset on subgrid land surface climate (2015-2100) from the Community Earth System Model. Geoscience Data Journal 10, 2 (2023), 208--219.
[30]
Yin Zhou and Oncel Tuzel. 2018. Voxelnet: End-to-end learning for point cloud based 3d object detection. In CVPR 2018. 4490--4499.
[31]
Johann Maximilian Zollner et al. 2024. Satellite Image Representations for Quantum Classifiers. Datenbank-Spektrum (2024), 1--9.

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
SIGSPATIAL '24: Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems
October 2024
743 pages
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: 22 November 2024

Check for updates

Author Tags

  1. Spatial representation learning
  2. location encoding
  3. polygon encoding
  4. spatially explicit artificial intelligence

Qualifiers

  • Short-paper
  • Research
  • Refereed limited

Funding Sources

Conference

SIGSPATIAL '24
Sponsor:

Acceptance Rates

SIGSPATIAL '24 Paper Acceptance Rate 37 of 122 submissions, 30%;
Overall Acceptance Rate 257 of 1,238 submissions, 21%

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • 0
    Total Citations
  • 210
    Total Downloads
  • Downloads (Last 12 months)210
  • Downloads (Last 6 weeks)147
Reflects downloads up to 14 Jan 2025

Other Metrics

Citations

View Options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Login options

Media

Figures

Other

Tables

Share

Share

Share this Publication link

Share on social media