On the use of recurrent neural networks for fast and accurate non-uniform gas radiation modeling

•Recurrent neural networks (RNN) are considered to treat radiative transfer in non-uniform gaseous media.•An accurate uniform model is presented.•Its extension to non-uniform media is detailed.•The corresponding non-uniform formulation is re-written in an RNN form.•The model is assessed against LBL calculations in high temperature scenarios.•The method bridges ideas introduced by Godson in the 50s with modern modeling strategies.