Gaussian Process Latent Variable Model

Gaussian Process Latent Variable Models (GPLVMs) are powerful unsupervised learning methods used for dimensionality reduction and related tasks like missing data imputation and signal separation. Current research focuses on improving scalability through techniques like stochastic variational inference and amortized inference, as well as addressing challenges such as model collapse and the development of more flexible kernel functions. These advancements enable GPLVMs to effectively analyze complex, high-dimensional datasets, particularly in fields like single-cell transcriptomics and spectroscopy, providing interpretable latent representations and improved performance compared to alternative approaches. The resulting insights are valuable for various scientific domains and practical applications requiring effective data analysis and understanding of underlying structures.

Papers

October 28, 2024

On Probabilistic Pullback Metrics on Latent Hyperbolic Manifolds
Luis Augenstein, Noémie Jaquier, Tamim Asfour, Leonel Rozo
Latent Space High Dimensional Data Riemannian Manifold Latent Manifold Probabilistic Metric Gaussian Process Latent Variable Model Hyperbolic Scaling

August 13, 2024

Variational Learning of Gaussian Process Latent Variable Models through Stochastic Gradient Annealed Importance Sampling
Jian Xu, Shian Du, Junmei Yang, Qianli Ma, Delu Zeng
Stochastic Gradient Importance Sampling Variational Learning Importance Weighting Annealed Importance Sampling Variational Bound Gaussian Process Latent Variable Model

May 6, 2024

Scalable Amortized GPLVMs for Single Cell Transcriptomics Data
Sarah Zhao, Aditya Ravuri, Vidhi Lalchand, Neil D. Lawrence
Variational Inference Potential Scalability Single Cell Single Cell RNA Gaussian Process Latent Variable Model

April 2, 2024

Preventing Model Collapse in Gaussian Process Latent Variable Models
Ying Li, Zhidi Lin, Feng Yin, Michael Minyi Zhang
Gaussian Process Variational Inference Latent Representation Model Collapse Gaussian Process Latent Variable Model

February 14, 2024

Mixed-Output Gaussian Process Latent Variable Models
James Odgers, Ruby Sedgwick, Chrysoula Kappatou, Ruth Misener, Sarah Filippi
Latent Variable Clean Speech Bayesian Nonparametric Gaussian Process Latent Variable Model

February 25, 2022

Generalised Gaussian Process Latent Variable Models (GPLVM) with Stochastic Variational Inference
Vidhi Lalchand, Aditya Ravuri, Neil D. Lawrence
Gaussian Process Latent Variable Variational Method Stochastic Variational Inference Gaussian Process Latent Variable Model