J-PAS (Javalambre Physics of the Accelerating Universe Astrophysical Survey) will present a groundbreaking photometric survey covering 8500 deg2 of the visible sky from Javalambre, capturing data in 56 narrow-band filters. This survey promises to revolutionise galaxy evolution studies by observing ∼108 galaxies with low spectral resolution. A crucial aspect of this analysis involves predicting stellar population parameters from the observed galaxy photometry. In this study, we combined the exquisite J-PAS photometry with state-of-the-art single stellar population (SSP) libraries to accurately predict stellar age, metallicity, and dust attenuation with a neural network (NN) model. The NN was trained on synthetic J-PAS photometry from different SSP libraries (E-MILES, Charlot & Bruzual, and XSL) to enhance the robustness of our predictions against individual SSP model variations and limitations. To create mock samples with varying observed magnitudes, we added artificial noise in the form of random Gaussian variations within typical observational uncertainties in each band. Our results indicate that the NN was able to accurately estimate stellar parameters for SSP models without any evident degeneracies, surpassing a Bayesian SED-fitting method on the same test set. We obtained the median bias, scatter, and the percentage of outliers: μ= (0.01 dex, 0.00 dex, 0.00 mag), σNMAD= (0.23 dex, 0.29 dex, 0.04 mag), fo= (17%, 24%, 1%) at i ∼ 17 mag for the age, metallicity and dust attenuation, respectively. The accuracy of the predictions is highly dependent on the signal-to-noise ratio (S/N) of the photometry, achieving robust predictions up to i ∼ 20 mag.