Multi-View Oriented GPLVM: Expressiveness and Efficiency
This work addresses efficiency and expressiveness challenges in multi-view data representation learning, offering a scalable solution with broad applications in domains like computer vision and bioinformatics, though it is incremental as it builds on existing MV-GPLVM frameworks.
The paper tackled the limited kernel expressiveness and low computational efficiency of multi-view Gaussian process latent variable models (MV-GPLVMs) by introducing a new duality between spectral density and kernel function, leading to a Next-Gen Spectral Mixture (NG-SM) kernel with random Fourier feature approximation, which consistently outperformed state-of-the-art models in learning meaningful latent representations across diverse datasets.
The multi-view Gaussian process latent variable model (MV-GPLVM) aims to learn a unified representation from multi-view data but is hindered by challenges such as limited kernel expressiveness and low computational efficiency. To overcome these issues, we first introduce a new duality between the spectral density and the kernel function. By modeling the spectral density with a bivariate Gaussian mixture, we then derive a generic and expressive kernel termed Next-Gen Spectral Mixture (NG-SM) for MV-GPLVMs. To address the inherent computational inefficiency of the NG-SM kernel, we propose a random Fourier feature approximation. Combined with a tailored reparameterization trick, this approximation enables scalable variational inference for both the model and the unified latent representations. Numerical evaluations across a diverse range of multi-view datasets demonstrate that our proposed method consistently outperforms state-of-the-art models in learning meaningful latent representations.