Login / Signup
Direction Matters: On the Implicit Bias of Stochastic Gradient Descent with Moderate Learning Rate.
Jingfeng Wu
Difan Zou
Vladimir Braverman
Quanquan Gu
Published in:
ICLR (2021)
Keyphrases
</>
learning rate
stochastic gradient descent
weight vector
convergence rate
step size
training speed
convergence speed
learning algorithm
least squares
loss function
random forests
matrix factorization
training algorithm
random forest
perceptron algorithm
online algorithms
objective function