arXiv:2010.08127cs.LG
The Deep Bootstrap Framework: Good Online Learners are Good Offline Generalizers
Early discovery. Linked by one identifiable commenter. This paper is outside the main ranking; a mention is not an endorsement.
Abstract
We propose a new framework for reasoning about generalization in deep learning. The core idea is to couple the Real World, where optimizers take stochastic gradient steps on the empirical loss, to an Ideal World, where optimizers take steps on the population loss. This leads to an alternate decomposition of test error into: (1) the Ideal World test error plus (2) the gap between the two worlds. If the gap (2) is universally small, this reduces the problem of generalization in offline learning to the problem of optimization in online learning. We then give empirical evidence that this gap between worlds can be small in realistic deep learning settings, in particular supervised image classification. For example, CNNs generalize better than MLPs on image distributions in the Real World, but this is "because" they optimize faster on the population loss in the Ideal World. This suggests our framework is a useful tool for understanding generalization in deep learning, and lays a foundation for future research in the area.
Note that we also have ImageNet experiments, with entirely real data (non-synthetic). See Section 4, and in particular Figure 3 of the full paper: https://arxiv.org/abs/2010.08127 To clarify some other comments on this post: In all settings, we compare "Real World" and "Ideal…
preetumView on HNParent context
This is the exact comparison we make in the paper. We have subsampled ImageNet experiments as well; see Figure 3 in the full paper: https://arxiv.org/abs/2010.08127
preetumView on HNParent context
We have ImageNet experiments in Section 4 of the full paper: https://arxiv.org/abs/2010.08127
preetumView on HNParent context