Seems like Bayesian learning is the go to method now for reducing over fitting in ML models, I've read Manning's NLP in action, Deep Learning with python, O'Reilly Deep Learning foundations, Hands on learning with sickit learn and tensorflow ( the most prolific ML book in history ) but sadly it uses the Tensorflow framework which is very manual and hard to work with compared to Pytorch, also Mathematics for machine learning, and 2 other books I forgot their names. But none of them really mentioned Bayesian learning as an alternative to classical machine learning models for halting over fitting.
Please help me if you know something