Siamese networks for Poincar\'e embeddings and the reconstruction of evolutionary trees

TL;DR

This paper introduces a novel method combining Siamese networks, Poincaré embeddings, and neighbor joining to reconstruct evolutionary trees from high-dimensional phenotypic data like bird song spectrograms, without predefined features.

Contribution

It presents a new approach that uses Siamese networks to learn embeddings solely from leaf nodes, improving phylogenetic inference from phenotypic traits.

Findings

Effective on synthetic data

Successful reconstruction of finch species trees

Outperforms traditional methods in accuracy

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: Poincar\'e embeddings for dimensionality reduction and distance computation, and the neighbor 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.