I have an input-output system, which is fully determined by 256 parameters, of which I know a significant amount are of less importance to the input-output pattern.
The data I have is some (64k in total) input-parameter-output match.
My goal is to compress these 256 parameters to a smaller scale (like 32) using an encoder of some kind while being able to preserve the response pattern.
But I can't seem to find a proper network for this particular problem, because I'm not trying to fit these parameters (they all have a mean of one and variance of 1/4), but rather its influence on the output, so traditional data-specific operations will not work in this case.
