Trained neural network to optimize input variable

Hi

I have a input matrix X = [x1, x2] with N rows of data and target matrix Y = [y1 y2] of same number of rows. I ran the data in a neural network to obtain a trained neural network function. Currently, I can provide some [x1, x2] value to attain [y1,y2] value for the input from the trained neural network function. However, I want to attain the optimized input x2 for a specific x1 value based on y1,y2 value. Range for y1 and y2 = [0 1], Range for x1 = [1200 2800] ans x2 = [0.05 0.95].

Can someone please help me with this?

Answer 1

[Greg Heath]

1. "OPTIMIZED INPUT" is ill-defined
2. Just train another net x2 = f2(x1,y1,y2)

Hope this helps,

THANK YOU FOR FORMALLY ACCEPTING MY ANSWER

GREG