3 ms·
The fair comparison is comparing the number of "free variables" in the system. This is how machine learning models are compared. You don't compare "number of fr
by argonaut 10y ago
The fair comparison is comparing the number of "free variables" in the system. This is how machine learning models are compared. You don't compare "number of free parameters in linear regression" with "number of layers of neural nets," you compare number of parameters with number of parameters.