Mechanistic mode connectivity
We study neural network loss landscapes through the lens of mode connectivity, the
observation that minimizers of neural networks retrieved via training on a dataset are …
observation that minimizers of neural networks retrieved via training on a dataset are …
Late stop**: Avoiding confidently learning from mislabeled examples
Sample selection is a prevalent method in learning with noisy labels, where small-loss data
are typically considered as correctly labeled data. However, this method may not effectively …
are typically considered as correctly labeled data. However, this method may not effectively …
Can neural network memorization be localized?
Memorization through the lens of curvature of loss function around samples
Deep neural networks are over-parameterized and easily overfit the datasets they train on.
In the extreme case, it has been shown that these networks can memorize a training set with …
In the extreme case, it has been shown that these networks can memorize a training set with …