![[2502.15110] Variational phylogenetic inference with products over bipartitions](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2502.15110] Variational phylogenetic inference with products over bipartitions
Summary
This paper presents a novel variational Bayesian method for inferring ultrametric phylogenetic trees, improving accuracy and efficiency in posterior distribution approximations without relying on MCMC techniques.
Why It Matters
Understanding evolutionary dynamics through Bayesian phylogenetics is crucial for various fields, including epidemiology and conservation biology. This research offers a more efficient approach to phylogenetic inference, which can enhance analyses of genomic data, particularly in the context of viral evolution.
Key Takeaways
- Introduces a variational Bayesian approach for ultrametric phylogenetic trees.
- Eliminates the need for MCMC subroutines, streamlining the inference process.
- Demonstrates competitive accuracy with fewer gradient evaluations compared to existing methods.
- Applies the method to genomic datasets, including SARS-CoV-2, showcasing its practical relevance.
- Offers a closed-form density for distributions over trees, enhancing computational efficiency.
Statistics > Machine Learning arXiv:2502.15110 (stat) COVID-19 e-print Important: e-prints posted on arXiv are not peer-reviewed by arXiv; they should not be relied upon without context to guide clinical practice or health-related behavior and should not be reported in news media as established information without consulting multiple experts in the field. [Submitted on 21 Feb 2025 (v1), last revised 12 Feb 2026 (this version, v2)] Title:Variational phylogenetic inference with products over bipartitions Authors:Evan Sidrow, Alexandre Bouchard-Côté, Lloyd T. Elliott View a PDF of the paper titled Variational phylogenetic inference with products over bipartitions, by Evan Sidrow and 2 other authors View PDF Abstract:Bayesian phylogenetics is vital for understanding evolutionary dynamics, and requires accurate and efficient approximation of posterior distributions over trees. In this work, we develop a variational Bayesian approach for ultrametric phylogenetic trees. We present a novel variational family based on coalescent times of a single-linkage clustering and derive a closed-form density for the resulting distribution over trees. Unlike existing methods for ultrametric trees, our method performs inference over all of tree space, it does not require any Markov chain Monte Carlo subroutines, and our variational family is differentiable. Through experiments on benchmark genomic datasets and an application to the viral RNA of SARS-CoV-2, we demonstrate that our method achieve...
