Hacker News new | ask | show | jobs
by irq11 2554 days ago
That is not even close to the same thing.

If you train an ensemble of models with random dropout, you have an ensemble. Models trained with dropout will still have significant variation from run to run.

1 comments

> That is not even close to the same thing.

It's a common interpretation: https://arxiv.org/abs/1706.06859

There may be a paper on it, but it’s not a common view.

In particular, this paper neglected to do the obvious thing: ensemble networks trained with dropout. It improves performance over dropout alone.