Adam's implicit bias depends on learning rate schedule
Adam maximizes weighted l2 margin for constant schedules on homogeneous models.
Stories per day · 30 days
1 stories, newest first · one card per story
1 stories, newest first · one card per story
Adam maximizes weighted l2 margin for constant schedules on homogeneous models.
Today 1.5K readers · 2.4K views · 757 reads · all time 189K views