Machine Learning General Truth – There should exist a truth structure common to all problems. Just because it will simplify a large universe. So when iterating a problem the ending result is a function like a*b+b*c = ans.The reason for the structure would be to get fast solutions
Machine Learning Physics Guess – If pi is the truth in a simple rotation. What is the truth loss function for a 3 body systems? Could there exist teleportation of energy since the planet system can not be disturbed. // Per
If there was an error you need energy.
Something I will test. Taylor series approximation of the function near each x label.
Food innovation – Try the set of possibilities. Fast Noodles + boiled water and later after 6 min add milk to cool it off and give it more flavor. // Per
Machine Learning – Hot encoding is not statistically ?friendly. I think you need the target to be a double normal distribution. Around 0 and around 1. For this you can add yt = yt +m ; m = model2(X) and then with subtract m from the model(X_test) – m.
Important Machine Learning Human Thinking – Move the body while your thinking. Move your arms, talk loud etc. Why? Since every mass unit in the body wants to improve the loss value set by the brain. There for everything adapts to find the answer. // Per
Machine Learning Guess – Could letting a layer act as incriment at the index i,j and multiply/add? with value for the next layer?
Guess – In machine learning math there is a connection between complex numbers and variation. So in F.sum(X) numbers. They to have variation in X. So there is a possibility for more variation in matrix multiplication other than F.sum() // Per