I am trying to understand what neural nets have in common with multiple linear regressions. Is is correct that every node in every layer (not counting the input layer) is just a multiple regression with the variables being the nodes in the preceding layer? I've included a picture to show what I mean. How is this understanding wrong/right?

enter image description here


marked as duplicate by Franck Dernoncourt, mdewey, kjetil b halvorsen, Community Mar 2 '17 at 19:37

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.