Evaluating the Stability of Deep Learning Latent Feature Spaces

arXiv:2402.11404v1 Announce Type: new
Abstract: High-dimensional datasets present substantial challenges in statistical modeling across various disciplines, necessitating effective dimensionality reduction methods. Deep learning approaches, notable for their capacity to distill essential features from complex data, facilitate modeling, visualization, and compression through reduced dimensionality latent feature spaces, have wide applications from bioinformatics to earth sciences. This study introduces a novel workflow to evaluate the stability of these latent spaces, ensuring consistency and reliability in subsequent analyses. Stability, defined as the invariance …

abstract applications arxiv bioinformatics capacity challenges compression cs.lg data datasets deep learning dimensionality feature features modeling spaces stability statistical statistical modeling through type visualization