1 paper
Rahul Vashisht, P. Krishna Kumar, Harsha Vardhan Govind +1
Neural networks trained with stochastic gradient descent exhibit an inductive bias towards simpler decision boundaries, typically converging to a narrow family of functions, and of…