Probabilistic Auto-Associative Models and Semi-Linear PCA
This work provides a domain-specific incremental extension of PCA for data analysis, particularly in fields like astronomy.
The authors tackled the problem of extending PCA to semi-linear auto-associative models by proposing a probabilistic Gaussian model, which they validated through numerical experiments on simulated and real astronomical datasets, showing its effectiveness.
Auto-Associative models cover a large class of methods used in data analysis. In this paper, we describe the generals properties of these models when the projection component is linear and we propose and test an easy to implement Probabilistic Semi-Linear Auto- Associative model in a Gaussian setting. We show it is a generalization of the PCA model to the semi-linear case. Numerical experiments on simulated datasets and a real astronomical application highlight the interest of this approach