all InfoSec news
Stability is Stable: Connections between Replicability, Privacy, and Adaptive Generalization. (arXiv:2303.12921v1 [cs.LG])
cs.CR updates on arXiv.org arxiv.org
The notion of replicable algorithms was introduced in Impagliazzo et al.
[STOC '22] to describe randomized algorithms that are stable under the
resampling of their inputs. More precisely, a replicable algorithm gives the
same output with high probability when its randomness is fixed and it is run on
a new i.i.d. sample drawn from the same distribution. Using replicable
algorithms for data analysis can facilitate the verification of published
results by ensuring that the results of an analysis will be …
algorithm algorithms analysis connections data data analysis distribution high inputs precisely privacy randomness run under verification