Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 147 tok/s
Gemini 2.5 Pro 52 tok/s Pro
GPT-5 Medium 27 tok/s Pro
GPT-5 High 30 tok/s Pro
GPT-4o 96 tok/s Pro
Kimi K2 188 tok/s Pro
GPT OSS 120B 398 tok/s Pro
Claude Sonnet 4.5 36 tok/s Pro
2000 character limit reached

What Do You See in Common? Learning Hierarchical Prototypes over Tree-of-Life to Discover Evolutionary Traits (2409.02335v2)

Published 3 Sep 2024 in cs.CV

Abstract: A grand challenge in biology is to discover evolutionary traits - features of organisms common to a group of species with a shared ancestor in the tree of life (also referred to as phylogenetic tree). With the growing availability of image repositories in biology, there is a tremendous opportunity to discover evolutionary traits directly from images in the form of a hierarchy of prototypes. However, current prototype-based methods are mostly designed to operate over a flat structure of classes and face several challenges in discovering hierarchical prototypes, including the issue of learning over-specific prototypes at internal nodes. To overcome these challenges, we introduce the framework of Hierarchy aligned Commonality through Prototypical Networks (HComP-Net). The key novelties in HComP-Net include a novel over-specificity loss to avoid learning over-specific prototypes, a novel discriminative loss to ensure prototypes at an internal node are absent in the contrasting set of species with different ancestry, and a novel masking module to allow for the exclusion of over-specific prototypes at higher levels of the tree without hampering classification performance. We empirically show that HComP-Net learns prototypes that are accurate, semantically consistent, and generalizable to unseen species in comparison to baselines.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (62)
  1. Complexity, evolvability, and the process of adaptation. Annual Review of Ecology, Evolution, and Systematics, 53, 2022.
  2. Morphobank: phylophenomics in the “cloud”. Cladistics, 27(5):529–537, 2011.
  3. Giant taxon-character matrices: quality of character constructions remains critical regardless of size. Cladistics, 33(2):198–219, 2017.
  4. Paul C Sereno. Logical basis for morphological characters in phylogenetics. Cladistics, 23(6):565–587, 2007.
  5. Computer vision, machine learning, and the promise of phenomics in ecology and evolutionary biology. Frontiers in Ecology and Evolution, 9:642774, 2021.
  6. The inaturalist species classification and detection dataset. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 8769–8778, 2018.
  7. A survey of digitized data from us fish collections in the idigbio data aggregator. PloS one, 13(12):e0207636, 2018.
  8. The caltech-ucsd birds-200-2011 dataset. 2011.
  9. Discovering novel biological traits from images using phylogeny-guided neural networks. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 3966–3978, 2023.
  10. This looks like that: deep learning for interpretable image recognition. Advances in neural information processing systems, 32, 2019.
  11. Protopshare: Prototypical parts sharing for similarity discovery in interpretable image classification. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 1420–1430, 2021.
  12. Interpretable image classification with differentiable prototypes assignment. In European Conference on Computer Vision, pages 351–368. Springer, 2022.
  13. Neural prototype trees for interpretable fine-grained image recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14933–14943, 2021.
  14. Early bursts of body size and shape evolution are rare in comparative data. Evolution, 64(8):2385–2396, 2010.
  15. Model adequacy and the macroevolution of angiosperm functional traits. The American Naturalist, 186(2):E33–E50, 2015.
  16. Heliconius collection (cambridge butterfly), 2024.
  17. Interpretable image recognition with hierarchical prototypes. In Proceedings of the AAAI Conference on Human Computation and Crowdsourcing, volume 7, pages 32–40, 2019.
  18. Pip-net: Patch-based intuitive prototypes for interpretable image classification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 2744–2753, 2023.
  19. This looks like that… does it? shortcomings of latent space prototype interpretability in deep networks. arXiv preprint arXiv:2105.02968, 2021.
  20. Hive: Evaluating the human interpretability of visual explanations. In European Conference on Computer Vision, pages 280–298. Springer, 2022.
  21. Understanding contrastive representation learning through alignment and uniformity on the hypersphere. In International Conference on Machine Learning, pages 9929–9939. PMLR, 2020.
  22. Representation learning via consistent assignment of views to clusters. In Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing, pages 987–994, 2022.
  23. Interpretable image recognition by constructing transparent embedding space. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 895–904, 2021.
  24. Orthogonal convolutional neural networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11505–11515, 2020.
  25. Categorical reparameterization with gumbel-softmax. arXiv preprint arXiv:1611.01144, 2016.
  26. The global diversity of birds in space and time. Nature, 491:444–448, 2012.
  27. Cambridge butterfly wing collection batch 10, November 2020.
  28. Sheffield butterfly wing collection - Patricio Salazar, Nicola Nadeau, Ikiam broods batch 1 and 2, November 2020.
  29. Cambridge butterfly wing collection batch 2, May 2019.
  30. Cambridge butterfly wing collection batch 3, May 2019.
  31. Cambridge butterfly wing collection batch 4, May 2019.
  32. Cambridge butterfly wing collection batch 5, May 2019.
  33. Miscellaneous Heliconius wing photographs (2001-2019) Part 1, February 2019.
  34. Miscellaneous Heliconius wing photographs (2001-2019) Part 3, February 2019.
  35. Cambridge butterfly wing collection batch 6, May 2019.
  36. Cambridge butterfly wing collection - Chris Jiggins 2001/2 broods batch 1, January 2019.
  37. Cambridge butterfly wing collection - Chris Jiggins 2001/2 broods batch 2, January 2019.
  38. Cambridge butterfly wing collection - Patricio Salazar PhD wild specimens batch 3, October 2020.
  39. Cambridge butterfly wing collection batch 1- version 2, May 2019.
  40. Cambridge and collaborators butterfly wing collection batch 10, May 2019.
  41. Cambridge butterfly wing collection - Patricio Salazar PhD wild and bred specimens batch 1, December 2018.
  42. Cambridge butterfly wing collection batch 7, May 2019.
  43. Cambridge butterfly wing collection - Patricio Salazar PhD wild and bred specimens batch 2, January 2019.
  44. Brazilian Butterflies Collected December 2020 to January 2021, February 2022.
  45. Cambridge butterfly wing collection batch 8, May 2019.
  46. Cambridge butterfly wing collection batch 9, May 2019.
  47. Cambridge butterfly collection - GMK Broods Ikiam 2018, November 2020.
  48. Heliconius erato cyrbia, Cook Islands (New Zealand) 2016, 2019, 2021, September 2021.
  49. Miscellaneous Heliconius wing photographs (2001-2019) Part 2, February 2019.
  50. Camilo Salazar and Cambridge butterfly wing collection batch 1, May 2019.
  51. University of Helsinki butterfly collection - Anniina Mattila bred specimens, February 2019.
  52. Open tree of life synthetic tree, 2019.
  53. rotl: an r package to interact with the open tree of life data. Methods in Ecology and Evolution, 7(12):1476–1481, 2016.
  54. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
  55. A simple interpretable transformer for fine-grained image classification and analysis. arXiv preprint arXiv:2311.04157, 2023.
  56. The dimensionality of genetic variation for wing shape in drosophila melanogaster. Evolution, 59(5):1027–1038, 2005.
  57. Hierarchical conditioning of diffusion models using tree-of-life for studying species evolution. arXiv preprint arXiv:2408.00160, 2024.
  58. Knowledge-guided machine learning: Current trends and future prospects. arXiv preprint arXiv:2403.15989, 2024.
  59. Linking of digital images to phylogenetic data matrices using a morphological ontology. Systematic Biology, 56(2):283–294, 2007.
  60. Abien Fred Agarap. Deep learning using rectified linear units (relu). arXiv preprint arXiv:1803.08375, 2018.
  61. Trivialaugment: Tuning-free yet state-of-the-art data augmentation. In Proceedings of the IEEE/CVF international conference on computer vision, pages 774–782, 2021.
  62. R. Farrell. Cub-200-2011 segmentations (1.0) [data set], 2024.

Summary

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

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

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up

Don't miss out on important new AI/ML research

See which papers are being discussed right now on X, Reddit, and more:

Explore Trending Papers

“Emergent Mind helps me see which AI papers have caught fire online.”

Philip

Philip

Creator, AI Explained on YouTube