Design of the neural network used in this work. Left Panel: Neural network used in the training stage. Φ and Ψ represent the encoder and decoder respectively. Tⁱ are the elements of the training set and are the elements of the anchor set. Φ(Tⁱ) and are the representations of Tⁱ and , respectively, in the encoder (feature) space. is the Euclidean barycentric representation of Φ(Tⁱ) in terms of d anchor points , which is fed to the decoder. Ψ(Θ) is the reconstructed output of the decoder. The network is trained by minimising the error between the input Tⁱ and output Ψ(Θ) temperature profiles. Right panel: Neural network (IAE model) of temperature profiles, where λ₁, …, λ_d are the input parameters and IAE([λ₁, …, λ_d]) is the output temperature profile. The decoder is not required in any step here.