Sample-compressible distribution families are shown to be PAC-learnable under additive noise and adversarial corruption, given new stability and low-frequency assumptions.
Private distribution learning with public data: The view from sample compression
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Robust Learnability of Sample-Compressible Distributions under Noisy or Adversarial Perturbations
Sample-compressible distribution families are shown to be PAC-learnable under additive noise and adversarial corruption, given new stability and low-frequency assumptions.