[P] Why does regularizing the bias lead to underfitting in neural networks?

Hi everyone!

I am a PhD student at the Indian Institute of Technology Kharagpur (IIT Kgp), working in the domain of understanding neural networks from a theoretical perspective. I am really passionate about DL and ML and wanted to give my own spin on many of the things I have learned while working through my Masters' and now PhD. I also get a lot of new ideas that aren't flashy enough to be put down in the form of a …

bias machinelearning networks neural networks underfitting