skip to main content
10.1145/3534678.3539448acmconferencesArticle/Chapter ViewAbstractPublication PageskddConference Proceedingsconference-collections
research-article

Robust Inverse Framework using Knowledge-guided Self-Supervised Learning: An application to Hydrology

Published: 14 August 2022 Publication History

Abstract

Machine Learning is beginning to provide state-of-the-art performance in a range of environmental applications such as streamflow prediction in a hydrologic basin. However, building accurate broad-scale models for streamflow remains challenging in practice due to the variability in the dominant hydrologic processes, which are best captured by sets of process-related basin characteristics. Existing basin characteristics suffer from noise and uncertainty, among many other things, which adversely impact model performance. To tackle the above challenges, in this paper, we propose a novel Knowledge-guided Self-Supervised Learning (KGSSL) inverse framework to extract system characteristics from driver(input) and response(output) data. This first-of-its-kind framework achieves robust performance even when characteristics are corrupted or missing. We evaluate the KGSSL framework in the context of stream flow modeling using CAMELS (Catchment Attributes and MEteorology for Large-sample Studies) which is a widely used hydrology benchmark dataset. Specifically, KGSSL outperforms baseline by 16% in predicting missing characteristics. Furthermore, in the context of forward modelling, KGSSL inferred characteristics provide a 35% improvement in performance over a standard baseline when the static characteristic are unknown.

Supplemental Material

MP4 File
Presentation video

References

[1]
Nans Addor et al. 2017. The CAMELS data set: Catchment attributes and meteorology for large-sample studies. Hydrology and Earth System Sciences (2017).
[2]
Camila Alvarez-Garreton et al. 2018. The CAMELS-CL dataset: Catchment attributes and meteorology for large sample studies-Chile dataset. Hydrology and Earth System Sciences (2018).
[3]
Lynton Ardizzone et al. 2018. Analyzing inverse problems with invertible neural networks. arXiv preprint arXiv:1808.04730 (2018).
[4]
Keith Beven. 2020. Deep learning, hydrological processes and the uniqueness of place. Hydrological Processes (2020).
[5]
Vinicius B.P. Chagas et al. 2020. CAMELS-BR: Hydrometeorological time series and landscape attributes for 897 catchments in Brazil. Earth System Science Data (2020).
[6]
Ting Chen et al. 2020. A simple framework for contrastive learning of visual representations. ICML (2020).
[7]
Gemma Coxon et al. 2020. CAMELS-GB: hydrometeorological time series and landscape attributes for 671 catchments in Great Britain. Earth System Science Data (2020).
[8]
Phuong D Dao et al. 2021. Improving hyperspectral image segmentation by applying inverse noise weighting and outlier removal for optimal scale selection. ISPRS Journal of Photogrammetry and Remote Sensing (2021).
[9]
Dapeng Feng et al. 2020. Enhancing streamflow forecast and extracting insights using long-short term memory networks with data integration at continental scales. Water Resources Research (2020).
[10]
Alexandre Ganachaud et al. 2000. Improved estimates of global ocean circulation, heat transport and mixing from hydrographic data. Nature (2000).
[11]
Spyros Gidaris et al. 2018. Unsupervised representation learning by predicting image rotations. arXiv preprint arXiv:1803.07728 (2018).
[12]
Zhen Hao et al. 2021. CCAM: China Catchment Attributes and Meteorology dataset. Earth System Science Data (2021).
[13]
Xiaowei Jia et al. 2019. Physics guided RNNs for modeling dynamical systems: A case study in simulating lake temperature profiles. SDM (2019).
[14]
Longlong Jing et al. 2020. Self-supervised visual feature learning with deep neural networks: A survey. IEEE TPAMI (2020).
[15]
Anuj Karpatne et al. 2017. Theory-guided data science: A new paradigm for scientific discovery from data. IEEE Transactions on knowledge and data engineering (2017).
[16]
Anuj Karpatne et al. 2022. Knowledge Guided Machine Learning: Accelerating Discovery using Scientific Knowledge and Data. CRC Press.
[17]
Ankush Khandelwal et al. 2020. Physics guided machine learning methods for hydrology. arXiv preprint arXiv:2012.02854 (2020).
[18]
Yuji Kim et al. 2018. Geophysical inversion versus machine learning in inverse problems. The Leading Edge (2018).
[19]
Frederik Kratzert et al. 2019. Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets. Hydrology and Earth System Sciences (2019).
[20]
Karthik Kumarasamy et al. 2018. Calibration parameter selection and watershed hydrology model evaluation in time and frequency domains. Water (2018).
[21]
Gustav Larsson et al. 2017. Colorization as a proxy task for visual understanding. CVPR (2017).
[22]
Wei-Chiu Ma et al. 2020. Deep feedback inverse problem solver. European Conference on Computer Vision (2020).
[23]
Andrew J Newman et al. 2015. Gridded ensemble precipitation and temperature estimates for the contiguous United States. Journal of Hydrometeorology (2015).
[24]
Brent D Newman et al. 2006. Ecohydrology of water-limited environments: A scientific vision. Water resources research (2006).
[25]
Mehdi Noroozi et al. 2016. Unsupervised learning of visual representations by solving jigsaw puzzles. ECCV (2016).
[26]
Gregory Ongie et al. 2020. Deep learning techniques for inverse problems in imaging. IEEE Journal on Selected Areas in Information Theory (2020).
[27]
Aaron van den Oord et al. 2018. Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748 (2018).
[28]
Deepak Pathak et al. 2016. Context encoders: Feature learning by inpainting. CVPR (2016).
[29]
Petr Pecha et al. 2021. Determination of radiological background fields designated for inverse modelling during atypical low wind speed meteorological episode. Atmospheric Environment (2021).
[30]
Sriram Ravula et al. 2021. Inverse Problems Leveraging Pre-trained Contrastive Representations. NeurIPS (2021).
[31]
Jordan S Read et al. 2019. Process-guided deep learning predictions of lake water temperature. Water Resources Research (2019).
[32]
Jeroen Ritsema et al. 2000. Seismic imaging of structural heterogeneity in Earth's mantle: evidence for large-scale mantle flow. Science Progress (1933-) (2000).
[33]
Ortal Senouf et al. 2019. Self-supervised learning of inverse problem solvers in medical imaging.
[34]
Risto Vuorio et al. 2019. Multimodal model-agnostic meta-learning via task-aware modulation. arXiv preprint arXiv:1910.13616 (2019).
[35]
Jared D Willard et al. 2022. Daily surface temperatures for 185,549 lakes in the conterminous United States estimated using deep learning (1980--2020). Limnology and Oceanography Letters (2022).
[36]
R Iestyn Woolway et al. 2021. Winter inverse lake stratification under historic and future climate change. Limnology and Oceanography Letters (2021).
[37]
Haiyan Zhou et al. 2014. Inverse methods in hydrogeology: Evolution and recent trends. Advances in Water Resources (2014).
[38]
Luisa Zintgraf et al. 2019. Fast context adaptation via meta-learning. ICML (2019).

Cited By

View all
  • (2024)Advancing Hydrology through Machine Learning: Insights, Challenges, and Future Directions Using the CAMELS, Caravan, GRDC, CHIRPS, PERSIANN, NLDAS, GLDAS, and GRACE DatasetsWater10.3390/w1613190416:13(1904)Online publication date: 3-Jul-2024
  • (2024)EvaNetProceedings of the Thirty-Third International Joint Conference on Artificial Intelligence10.24963/ijcai.2024/133(1200-1208)Online publication date: 3-Aug-2024
  • (2024)Knowledge-guided machine learning can improve carbon cycle quantification in agroecosystemsNature Communications10.1038/s41467-023-43860-515:1Online publication date: 8-Jan-2024
  • Show More Cited By

Index Terms

  1. Robust Inverse Framework using Knowledge-guided Self-Supervised Learning: An application to Hydrology

      Recommendations

      Comments

      Information & Contributors

      Information

      Published In

      cover image ACM Conferences
      KDD '22: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
      August 2022
      5033 pages
      ISBN:9781450393850
      DOI:10.1145/3534678
      Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

      Sponsors

      Publisher

      Association for Computing Machinery

      New York, NY, United States

      Publication History

      Published: 14 August 2022

      Permissions

      Request permissions for this article.

      Check for updates

      Author Tags

      1. forward modeling
      2. inverse modeling
      3. self-supervised learning

      Qualifiers

      • Research-article

      Conference

      KDD '22
      Sponsor:

      Acceptance Rates

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

      Upcoming Conference

      KDD '25

      Contributors

      Other Metrics

      Bibliometrics & Citations

      Bibliometrics

      Article Metrics

      • Downloads (Last 12 months)97
      • Downloads (Last 6 weeks)14
      Reflects downloads up to 20 Feb 2025

      Other Metrics

      Citations

      Cited By

      View all
      • (2024)Advancing Hydrology through Machine Learning: Insights, Challenges, and Future Directions Using the CAMELS, Caravan, GRDC, CHIRPS, PERSIANN, NLDAS, GLDAS, and GRACE DatasetsWater10.3390/w1613190416:13(1904)Online publication date: 3-Jul-2024
      • (2024)EvaNetProceedings of the Thirty-Third International Joint Conference on Artificial Intelligence10.24963/ijcai.2024/133(1200-1208)Online publication date: 3-Aug-2024
      • (2024)Knowledge-guided machine learning can improve carbon cycle quantification in agroecosystemsNature Communications10.1038/s41467-023-43860-515:1Online publication date: 8-Jan-2024
      • (2024)Using Knowledge-Guided Machine Learning To Assess Patterns of Areal Change in Waterbodies across the Contiguous United StatesEnvironmental Science & Technology10.1021/acs.est.3c0578458:11(5003-5013)Online publication date: 6-Mar-2024
      • (2024)Predicting the growth trajectory and yield of greenhouse strawberries based on knowledge-guided computer visionComputers and Electronics in Agriculture10.1016/j.compag.2024.108911220:COnline publication date: 17-Jul-2024
      • (2024)Evolution-Based Feature Selection for Predicting Dissolved Oxygen Concentrations in LakesParallel Problem Solving from Nature – PPSN XVIII10.1007/978-3-031-70085-9_25(398-415)Online publication date: 7-Sep-2024
      • (2023)Cyberinfrastructure for sustainability sciencesEnvironmental Research Letters10.1088/1748-9326/acd9dd18:7(075002)Online publication date: 7-Jul-2023
      • (2022)Meta-Transfer Learning: An application to Streamflow modeling in River-streams2022 IEEE International Conference on Data Mining (ICDM)10.1109/ICDM54844.2022.00026(161-170)Online publication date: Nov-2022

      View Options

      Login options

      View options

      PDF

      View or Download as a PDF file.

      PDF

      eReader

      View online with eReader.

      eReader

      Figures

      Tables

      Media

      Share

      Share

      Share this Publication link

      Share on social media