Minimum Description Length Principle in Supervised Learning with Application to Lasso | Read Paper on Bytez