Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
133 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

LD-SDM: Language-Driven Hierarchical Species Distribution Modeling (2312.08334v1)

Published 13 Dec 2023 in cs.CV

Abstract: We focus on the problem of species distribution modeling using global-scale presence-only data. Most previous studies have mapped the range of a given species using geographical and environmental features alone. To capture a stronger implicit relationship between species, we encode the taxonomic hierarchy of species using a LLM. This enables range mapping for any taxonomic rank and unseen species without additional supervision. Further, we propose a novel proximity-aware evaluation metric that enables evaluating species distribution models using any pixel-level representation of ground-truth species range map. The proposed metric penalizes the predictions of a model based on its proximity to the ground truth. We describe the effectiveness of our model by systematically evaluating on the task of species range prediction, zero-shot prediction and geo-feature regression against the state-of-the-art. Results show our model outperforms the strong baselines when trained with a variety of multi-label learning losses.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (49)
  1. Palm 2 technical report. arXiv preprint arXiv:2305.10403, 2023.
  2. Geography-aware self-supervised learning. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 10181–10190, 2021.
  3. A near-linear time algorithm for the chamfer distance. arXiv preprint arXiv:2307.03043, 2023.
  4. Species distribution modeling for machine learning practitioners: A review. In ACM SIGCAS conference on computing and sustainable societies, pages 329–348, 2021.
  5. Spherical fourier neural operators: Learning stable dynamics on the sphere. International conference on machine learning, 2023.
  6. A deep learning approach to species distribution modelling. Multimedia Tools and Applications for Environmental & Biodiversity Informatics, pages 169–199, 2018.
  7. Overview of geolifeclef 2019: plant species prediction using environment and animal occurrences. In CLEF 2019 Working Notes-Conference and Labs of the Evaluation Forum, page 257, 2019.
  8. Jointly estimating spatial sampling effort and habitat suitability for multiple species from opportunistic presence-only data. Methods in Ecology and Evolution, 12(5):933–945, 2021.
  9. From individual to joint species distribution models: A comparison of model complexity and predictive performance. Journal of Biogeography, 46(10):2260–2274, 2019.
  10. Bias reduction via end-to-end shift learning: Application to citizen science. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 493–500, 2019.
  11. Deep multi-species embedding. arXiv preprint arXiv:1609.09353, 2016.
  12. Geo-aware networks for fine-grained recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops, pages 0–0, 2019.
  13. The geolifeclef 2020 dataset. arXiv preprint arXiv:2004.04192, 2020.
  14. Multi-label learning from single positive labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 933–942, 2021.
  15. Spatial implicit neural representations for global-scale species mapping. International conference on machine learning, 2023.
  16. Species distribution models: ecological explanation and prediction across space and time. Annual review of ecology, evolution, and systematics, 40:677–697, 2009.
  17. Novel methods improve prediction of species’ distributions from occurrence data. Ecography, 29(2):129–151, 2006.
  18. Worldclim 2: new 1-km spatial resolution climate surfaces for global land areas. International journal of climatology, 37(12):4302–4315, 2017.
  19. Spatiotemporal exploratory models for broad-scale survey data. Ecological Applications, 20(8):2131–2147, 2010.
  20. Cascaded neural networks improving fish species prediction accuracy: the role of the biotic information. Scientific Reports, 8(1):4581, 2018.
  21. Asymmetric polynomial loss for multi-label classification. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE, 2023.
  22. Estimating species distributions from spatially biased citizen science data. Ecological Modelling, 422:108927, 2020.
  23. Evaluation metrics and validation of presence-only species distribution models based on distributional maps with varying coverage. Scientific Reports, 11(1):1482, 2021.
  24. Active learning-based species range estimation. Advances in Neural Information Processing Systems, 2023.
  25. Overview of geolifeclef 2021: Predicting species distribution from 2 million remote sensing images. In CLEF (Working Notes), pages 1451–1462, 2021.
  26. Overview of geolifeclef 2022: Predicting species presence from multi-modal remote sensing, bioclimatic and pedologic data. In CLEF 2022-Conference and Labs of the Evaluation Forum, pages 1940–1956, 2022.
  27. Presence-only geographical priors for fine-grained image classification. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 9596–9606, 2019.
  28. DARRYL I MacKENZIE. What are the issues with presence-absence data for wildlife managers? The Journal of Wildlife Management, 69(3):849–860, 2005.
  29. A comprehensive evaluation of predictive performance of 33 species distribution models at species and community levels. Ecological monographs, 89(3):e01370, 2019.
  30. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023.
  31. Robust asymmetric loss for multi-label long-tailed learning. arXiv preprint arXiv:2308.05542, 2023.
  32. Asymmetric loss for multi-label classification. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 82–91, 2021.
  33. The gbif integrated publishing toolkit: facilitating the efficient publishing of biodiversity data on the internet. PloS one, 9(8):e102623, 2014.
  34. Geographic location encoding with spherical harmonics and sinusoidal representation networks. arXiv preprint arXiv:2310.06743, 2023.
  35. Birdsat: Cross-view contrastive masked autoencoders for bird species classification and mapping. IEEE/CVF Winter Conference on Applications of Computer Vision, 2024.
  36. An evaluation of methods for modelling species distributions. Journal of biogeography, 31(10):1555–1568, 2004.
  37. ebird: A citizen-based bird observation network in the biological sciences. Biological conservation, 142(10):2282–2292, 2009.
  38. Multi-entity dependence learning with rich context via conditional variational auto-encoder. In Proceedings of the AAAI conference on artificial intelligence, 2018.
  39. Satbird: Bird species distribution modeling with remote sensing and citizen science data. Advances in Neural Information Processing Systems, 2023a.
  40. Bird distribution modelling using remote sensing and citizen science data. arXiv preprint arXiv:2305.01079, 2023b.
  41. Joint dynamic species distribution models: a tool for community ordination and spatio-temporal monitoring. Global Ecology and Biogeography, 25(9):1144–1158, 2016.
  42. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288, 2023.
  43. Predictive performance of presence-only species distribution models: a benchmark study with reproducible code. Ecological Monographs, 92(1):e01486, 2022.
  44. The inaturalist species classification and detection dataset. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 8769–8778, 2018.
  45. Peter D Wilson. Distance-based methods for the analysis of maps produced by species distribution models. Methods in Ecology and Evolution, 2(6):623–633, 2011.
  46. Density-aware chamfer distance as a comprehensive metric for point cloud completion. Advances in Neural Information Processing Systems, 2021.
  47. Segformer: Simple and efficient design for semantic segmentation with transformers. Advances in Neural Information Processing Systems, 34:12077–12090, 2021.
  48. Simple and robust loss design for multi-label learning with missing labels. arXiv preprint arXiv:2112.07368, 2021.
  49. Acknowledging the unknown for multi-label learning with single positive labels. In European Conference on Computer Vision, pages 423–440. Springer, 2022.
Citations (4)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now