Trained ResNets on CIFAR-10 retain measurable dependence on initialization scale under low-LR SGD (26.5 pp test accuracy spread) but not under Adam, indicating that practical inductive bias is shaped by the forgetting time scale of the optimizer and regularizers.
On the Trajectories of SGD Without Replacement
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
Mini-batch noise reverses how Adam's β2 controls anti-regularization, making default momentum values suitable for small batches but requiring β1 closer to β2 for large batches to favor flatter minima.
citing papers explorer
-
Do Deep Networks Forget Initialization? A Forgetting-Time View of Practical Inductive Bias
Trained ResNets on CIFAR-10 retain measurable dependence on initialization scale under low-LR SGD (26.5 pp test accuracy spread) but not under Adam, indicating that practical inductive bias is shaped by the forgetting time scale of the optimizer and regularizers.
-
The Effect of Mini-Batch Noise on the Implicit Bias of Adam
Mini-batch noise reverses how Adam's β2 controls anti-regularization, making default momentum values suitable for small batches but requiring β1 closer to β2 for large batches to favor flatter minima.
- Convergence of difference inclusions: a diameter criterion and step-size conditions