I find it is a failure of deep learning that a model is unable to spot these by itself. A dense layer would have a neuron connected to both f527 and f528. A pos

I find it is a failure of deep learning that a model is unable to spot these by itself. A dense layer would have a neuron connected to both f527 and f528. A positive weight on the first and a negative one on the second would give f527-f528. The fact that stochastic search is incapable of making these things emerge is problematic.