Model design of our algorithm (NEAT-VAE). (1) In the forward model, we assumed the 35 Galactic all-sky maps D to be generated by a smaller number of features Z and some additive noise N. The generative process f_θ, which we approximated by the decoder neural network, was learned by the algorithm. (2) We calculated the joint posterior distribution of the features Z, the network parameters θ, and the noise parameter ξ_N using statistical priors and Bayes’ theorem. (3) We approximated the posterior distribution using variational inference, the maximum entropy principle, and inverse transform sampling. Batches of spatially independent pixel vectors d_i served as input data, where each vector contains the spectral information of the same pixel in 35 Galactic all-sky maps. The algorithm was constructed to infer a latent representation z_i of the input data (encoder), and to regenerate its input d_i as accurate as possible from the latent space (decoder). The minimization objective, or loss function, guiding the algorithm’s learning process is Eq. (8).