Hostname: page-component-76c49bb84f-sz5hq Total loading time: 0 Render date: 2025-07-13T19:47:32.511Z Has data issue: false hasContentIssue false
  • English
  • Français

SIAMESE NETWORKS FOR POINCARÉ EMBEDDINGS AND THE RECONSTRUCTION OF EVOLUTIONARY TREES

Part of: Special Collection in honour of Professor Bernd Krauskopf Real and complex geometry

Published online by Cambridge University Press:  07 July 2025

CIRO CARVALLO Open the ORCID record for CIRO CARVALLO [Opens in a new window] ,
HERNÁN BOCACCIO Open the ORCID record for HERNÁN BOCACCIO [Opens in a new window] ,
GABRIEL B. MINDLIN Open the ORCID record for GABRIEL B. MINDLIN [Opens in a new window]  and
PABLO GROISMAN Open the ORCID record for PABLO GROISMAN [Opens in a new window]
CIRO CARVALLO
Affiliation:
Departamento de Matemática, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina; e-mail: ccarvallo@dm.uba.ar
HERNÁN BOCACCIO
Affiliation:
Departamento de Física, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires y CONICET – Universidad de Buenos Aires, Instituto de Física Interdisciplinaria y Aplicada (INFINA), Ciudad Universitaria, Buenos Aires, Argentina; e-mail: hbocaccio@gmail.com, gabo@df.uba.ar
GABRIEL B. MINDLIN
Affiliation:
Departamento de Física, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires y CONICET – Universidad de Buenos Aires, Instituto de Física Interdisciplinaria y Aplicada (INFINA), Ciudad Universitaria, Buenos Aires, Argentina; e-mail: hbocaccio@gmail.com, gabo@df.uba.ar
PABLO GROISMAN*
Affiliation:
Departamento de Matemática, Facultad de Ciencias Exactas y Naturales, https://ror.org/0081fs513 Universidad de Buenos Aires y CONICET – Universidad de Buenos Aires , Instituto de Matemática Luis A. Santaló (IMAS), Ciudad Universitaria, Buenos Aires, Argentina
*
e-mail: pgroisma@dm.uba.ar
Get access Rights & Permissions [Opens in a new window]

Abstract

We present a method for reconstructing evolutionary trees from high-dimensional data, with a specific application to bird song spectrograms. We address the challenge of inferring phylogenetic relationships from phenotypic traits, like vocalizations, without predefined acoustic properties. Our approach combines two main components: Poincaré embeddings for dimensionality reduction and distance computation, and the neighbour-joining algorithm for tree reconstruction. Unlike previous work, we employ Siamese networks to learn embeddings from only leaf node samples of the latent tree. We demonstrate our method’s effectiveness on both synthetic data and spectrograms from six species of finches.

Keywords

dynamical systems reconstructionphylogenetic treesneighbour-joining algorithmPoincaré embeddings

MSC classification

Secondary: 92D15: Problems related to evolution 51M10: Hyperbolic and elliptic geometries (general) and generalizations

Information

Type
Research Article
Information
The ANZIAM Journal , Volume 67 , 2025 , e25
Check for updates
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Australian Mathematical Publishing Association Inc

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

References

Atigh, M. G., Keller-Ressel, M. and Mettes, P., “Hyperbolic Busemann learning with ideal prototypes”, Adv. Neural Inf. Process Syst. 6 (2021) 103–115; https://proceedings.neurips.cc/paper˙files/paper/2021/file/01259a0cb2431834302abe2df60a1327-Paper.pdf.Google Scholar
Beecher, M. D. and Brenowitz, E. A., “Functional aspects of song learning in songbirds”, Trends Ecol. Evolut. 20 (2005) 143–149; doi:10.1016/j.tree.2005.01.00.CrossRefGoogle Scholar
Bistel, R., Martinez, A. and Mindlin, G. B., “An analysis of the persistence of Zonotrichia capensis themes using dynamical systems and machine learning tools”, Chaos Solitons Fractals 165 (2022) Article no. 112803; doi:10.1016/j.chaos.2022.112803.CrossRefGoogle Scholar
Boguna, M., Krioukov, D. and Klaffy, K., “Navigability of complex networks”, Nat. Phys. 5 (2009) 74–80; doi:10.1038/nphys1130.CrossRefGoogle Scholar
Bojanowski, P., Grave, E., Joulin, A. and Mikolov, T., “Enriching word vectors with subword information”, Trans. Associat. Comput. Linguist. 5 (2017) 135–146; doi:10.1162/tacl_a_00051.CrossRefGoogle Scholar
Bromley, J., Guyon, I., LeCun, Y., Säckinger, E. and Shah, R., “Signature verification using a “Siamese” time delay neural network”, Adv. Neural Inf. Process. Syst. 6 (1994) 737–744; https://proceedings.neurips.cc/paper˙files/paper/1993/file/288cc0ff022877bd3df94bc9360b9c5d-Paper.pdf.Google Scholar
Cannon, J. W., Floyd, W. J., Kenyon, R. and Parry, W. R., “Hyperbolic geometry”, in: Flavors of geometry, Vol. 22 (MSRI Publications, 1997) 59–115; https://www.math.ucdavis.edu/kapovich/RFG/cannon.pdf.10.1017/9781009701853.003CrossRefGoogle Scholar
Cate, C. T., “Birdsong and evolution”, in: Nature’s music (eds. P. Marler and H. Slabbekoorn) (Academic Press, San Diego, CA, 2004) 296–317; https://www.sciencedirect.com/science/article/pii/B978012473070050013X.CrossRefGoogle Scholar
Chami, I., Ying, R., Ré, C. and Leskovec, J., “Hyperbolic graph convolutional neural networks”, in: Advances in Neural Information Processing Systems (eds. H. Wallach, H. Larochelle, A. Beygelzimer, F. D. Alché-Bucand, E. Fox and R. Garnett) (Curran Associates Inc., 2019); https://proceedings.neurips.cc/paper_files/paper/2019/file/0415740eaa4d9decbc8da001d3fd805f-Paper.pdf.Google Scholar
Chen, Z. and Wiens, J., “The origins of acoustic communication in vertebrates”, Nat. Commun. 11 (2020) Article ID: 369; doi:10.1038/s41467-020-14356-3.Google Scholar
Chicco, D., “Siamese neural networks: an overview”, Artif. Neural Netw. 2190 (2020) 73–94; doi:10.1007/978-1-0716-0826-5_3.CrossRefGoogle Scholar
Chopra, S., Hadsell, R. and LeCun, Y., “Learning a similarity metric discriminatively, with application to face verification”, in: 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Volume 1 (IEEE, San Diego, CA, 2005) 539–546; doi:10.1109/CVPR.2005.202.CrossRefGoogle Scholar
Compeau, P. and Pevzner, P., Bioinformatics algorithms: an active learning approach, Volume 1, 2nd edn (Active Learning Publisher, 2015).Google Scholar
Farris, J. S., “Estimating phylogenetic trees from distance matrices”, Amer. Natur. 106(951) (1972) 645–668; http://www.jstor.org/stable/2459725.10.1086/282802CrossRefGoogle Scholar
Ge, S., Mishra, S., Kornblith, S., Li, C.-L. and Jacobs, D., “Hyperbolic contrastive learning for visual representations beyond objects”, in: IEEE/CVF Conference on Computer Vision and Pattern Recognition, Volume 12 (IEEE, Vancouver, BC, 2022) 6840–6849; doi:10.1109/CVPR52729.2023.00661.Google Scholar
GhadimiAtigh, M., Schoep, J., Acar, E., van Noord, N. and Mettes, P., “Hyperbolic image segmentation”, in: IEEE/CVF Conference on Computer Vision and Pattern Recognition, Volume 3 (IEEE, New Orleans, LA, 2022) 4453–4462; doi:10.1109/CVPR52688.2022.00441.Google Scholar
Greenberg, M. J., Euclidean and non-Euclidean geometries: development and history, 4th edn (W. H. Freeman, New York, 2008).Google Scholar
Guo, Y., Wang, X., Chen, Y. and Yu, S. X., “Clipped hyperbolic classifiers are super-hyperbolic classifiers”, in: IEEE/CVF Conference on Computer Vision and Pattern Recognition (IEEE, New Orleans, LA, 2022); doi:10.1109/CVPR52688.2022.00010.Google Scholar
Kenton Jacob Devlin, M.-W. C. and Toutanova, L. K., “BERT: Pre-training of deep bidirectional transformers for language understanding”, in: Proceedings of NAACL-HLT, Volumes 1–2 (eds. J. Burstein, C. Doran and T. Solorio) (Association for Computational Linguistics, Minneapolis, MN, 2019) 4171–4186; doi:10.18653/v1/N19-1423.Google Scholar
Khrulkov, V., Mirvakhabova, L., Ustinova, E., Oseledets, I. and Lempitsky, V., “Hyperbolic image embeddings”, in: IEEE/CVF Conference on Computer Vision and Pattern Recognition, Volume 4 (IEEE, Seattle, WA, 2019); doi:10.1109/CVPR42600.2020.00645.Google Scholar
Kingma, D. P. and Ba, J., “Adam: A method for stochastic optimization”, in: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, May 7–9, 2015, Conference Track Proceedings (2015); https://arxiv.org/abs/1412.6980.Google Scholar
Krioukov, D., Papadopoulos, F., Kitsak, M., Vahdat, A. and Boguna, M., “Hyperbolic geometry of complex networks”, Phys. Rev. E 82(3) (2010) 6; doi:10.1103/PhysRevE.82.036106.CrossRefGoogle Scholar
Lin, F.-Y., Bai, B., Bai, K., Ren, Y., Zhao, P. and Xu, Z., “Contrastive multi-view hyperbolic hierarchical clustering”, in: International Joint Conference on Artificial Intelligence (2022); https://arxiv.org/abs/2205.02618.Google Scholar
Liu, Q., Nickel, M. and Kiela, D., “Hyperbolic graph neural networks”, in: Advances on Neural Information Processing Systems (eds. H. Wallach, H. Larochelle, A. Beygelzimer and F. D. Alché-Buc, E. Fox and R. Garnett) (Curran Associates, Inc. 2019); https://proceedings.neurips.cc/paper˙files/paper/2019/file/103303dd56a731e377d01f6a37badae3-Paper.pdf.Google Scholar
Liu, S., Chen, J., Pan, L., Ngo, C.-W., Chua, T.-S. and Jiang, Y.-G., “Hyperbolic visual embedding learning for zero-shot recognition”, in: IEEE/CVF Conference on Computer Vision and Pattern Recognition (IEEE, Seattle, WA, 2020); doi:10.1109/CVPR42600.2020.00929.CrossRefGoogle Scholar
Liu, Z., Lin, W., Shi, Y. and Zhao, J., “A robustly optimized BERT pre-training approach with post-training”, in: Chinese Computational Linguistics (eds. S. Li, M. Sun, Y. Liu, H. Wu, L. Kang, W. Che, S. He and G. Rao) (Springer International Publishing, Cham, 2021) 471–484; doi:10.1007/978-3-030-84186-7_31.CrossRefGoogle Scholar
Martens, J., “Vocalizations and speciation of palearctic birds”, Ecol. Evolut. Acoust. Commun. Birds 1 (1996) 221–240; doi:10.7591/9781501736957-019.Google Scholar
Mason, N., Burns, K., Tobias, J., Claramunt, S., Seddon, N. and Derryberry, E., “Song evolution, speciation, and vocal learning in passerine birds”, Evolution 71 (2016) 786–796. doi:10.1111/evo.13159.CrossRefGoogle Scholar
Mathieu, E., Lan, C. L., Maddison, C. J., Tomioka, R. and Teh, Y. W., “Continuous hierarchical representations with Poincaré variational auto-encoders”, in: Advances in Neural Information Processing Systems, Volume 1 (2019); https://proceedings.neurips.cc/paper_files/paper/2019/file/0ec04cb3912c4f08874dd03716f80df1-Paper.pdf.Google Scholar
Mikolov, T., Chen, K., Corrado, G. and Dean, J., “Distributed representations of words and phrases and their compositionality”, in: Advances in Neural Information Processing Systems (Curran Associates, Inc. San Francisco, CA, 2013); https://proceedings.neurips.cc/paper˙files/paper/2013/file/9aa42b31882ec039965f3c4923ce901b-Paper.pdf.Google Scholar
Mikolov, T., Chen, K., Corrado, G. and Dean, J., “Efficient estimation of word representations in vector space”, in: 1st International Conference on Learning Representations, ICLR 2013, Scottsdale, AZ, May 2–4, 2013, Workshop Track Proceedings (IEEE, 2013); https://api.semanticscholar.org/CorpusID:5959482.Google Scholar
Moreira, G., Marques, M., Costeira, J. P. and Hauptmann, A. G., “Hyperbolic vs Euclidean embeddings in few-shot learning: two sides of the same coin”, in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (2024) 2082–2090; doi:10.1109/WACV57701.2024.00208.Google Scholar
Nagano, Y., Yamaguchi, S., Fujita, Y. and Koyama, M., “A wrapped normal distribution on hyperbolic space for gradient-based learning”, Int. Conf. Machine Learning 2 (2019) 4693–4702; https://proceedings.mlr.press/v97/nagano19a.html.Google Scholar
Nickel, M. and Kiela, D., “Poincaré embeddings for learning hierarchical representations”, in: Advances in Neural Information Processing Systems, (eds. I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan and R. Garnett) Volume 30 (Curran Associates, Inc. 2017); https://proceedings.neurips.cc/paper_files/paper/2017/file/59dfa2df42d9e3d41f5b02bfc32229dd-Paper.pdf.Google Scholar
Pennington, J., Socher, R. and Manning, C. D., “Glove: global vectors for word representation”, in: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) (eds. A. Moschitti, B. Pang and W. Daelemans) (Association for Computational Linguistics, Doha, 2014) 1532–1543; doi:10.3115/v1/D14-1162.CrossRefGoogle Scholar
Peters, M. E., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K. and Zettlemoyer, L., “Deep contextualized word representations”, in: Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) (eds. M. Walker, H. Ji and A. Stent) (Association for Computational Linguistics, New Orleans, Louisiana, 2018) 2227–2237; https://aclanthology.org/N18-1202/.Google Scholar
Price, J. J. and Lanyon, S., “Reconstructing the evolution of complex bird song in the oropendolas”, Evolution 56 (2002) 1514–1529; doi:10.1111/j.0014-3820.2002.tb01462.x.Google Scholar
Radford, A., Narasimhan, K., Salimans, T. and Sutskever, I., “Improving language understanding by generative pre-training”, OpenAI Preprint (2018) 1–12; https://cdn.openai.com/research-covers/language-unsupervised/language˙understanding˙paper.pdf.Google Scholar
Rivera, M., Edwards, J., Hauber, M. and Woolley, S., “Machine learning and statistical classification of birdsong link vocal acoustic features with phylogeny”, Sci. Rep. 13 (2023) Article no. 7076; doi:10.1038/s41598-023-33825-5.CrossRefGoogle Scholar
Saitou, N. and Nei, M., “The neighbor-joining method: a new method for reconstructing phylogenetic trees”, Mol. Biol. Evolut. 4(4) (1987) 406–425; doi:10.1093/oxfordjournals.molbev.a040454.Google Scholar
Sala, F., De Sa, C., Gu, A. and Ré, C., “Representation tradeoffs for hyperbolic embeddings”, Int. Conf. Mach. Learn. 80 (2018) 4460–4469; https://proceedings.mlr.press/v80/sala18a.html.Google Scholar
Tachibana, R. O., Oosugi, N. and Okanoya, K., “Semi-automatic classification of birdsong elements using a linear support vector machine”, PLoS One 9(3) (2014) 1–8; doi:10.1371/journal.pone.0092584.CrossRefGoogle Scholar
Taigman, Y., Yang, M., Ranzato, M. and Wolf, L., “Deepface: closing the gap to human-level performance in face verification”, in: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (IEEE, Ohio, 2014) 1701–1708; doi:10.1109/CVPR.2014.220.Google Scholar
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L. and Polosukhin, I., “Attention is all you need”, in: Advances in Neural Information Processing Systems, Volume 30 (31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, 2017); https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf.Google Scholar
Wimberger, P. H. and de Queiroz, A., “Comparing behavioral and morphological characters as indicators of phylogeny”, in: Phylogenies and the Comparative Method in Animal Behavior, (ed. E. P. Martins) Volume 4 (Oxford University Press, New York, 1996); doi:10.1093/oso/9780195092103.003.0007.CrossRefGoogle Scholar
Xeno-canto Foundation and N. B. Center, “Xeno-canto”; https://xeno-canto.org/.Google Scholar
Yue, Y., Lin, F., Yamada, K. D. and Zhang, Z., “Hyperbolic contrastive learning”, Preprint, 2023, arXiv:2302.01409.Google Scholar