Algorithmic robustness

cosmos 9th November 2018 at 12:01am
Generalization

robustness: the property that if a testing sample is "similar" to a training sample, then the testing error is close to the training error.

Robustness and generalization

similar to Algorithmic stability

See also here

Uses Lipschitz continuity

simply bounding the Lipschitz constant of the network is not enough to get a reasonable capacity control and the capacity bounds of the previous Section are not merely a consequence of bounding the Lipschitz constant.

Distance-based classification with lipschitz functions