Open AccessBook
An Introduction to Computational Learning Theory
Michael Kearns,Umesh Vazirani +1 more
Reads0
Chats0
TLDR
The probably approximately correct learning model Occam's razor the Vapnik-Chervonenkis dimension weak and strong learning learning in the presence of noise inherent unpredictability reducibility in PAC learning learning finite automata is described.Abstract:
The probably approximately correct learning model Occam's razor the Vapnik-Chervonenkis dimension weak and strong learning learning in the presence of noise inherent unpredictability reducibility in PAC learning learning finite automata by experimentation appendix - some tools for probabilistic analysis.read more
Citations
More filters
Book ChapterDOI
Selective sampling methods in one-class classification problems
Piotr Juszczak,Robert P. W. Duin +1 more
TL;DR: The goal of this paper is to show why the best or most often used selective sampling methods for two- or multi-class problems are not necessarily the best ones for the one-class classification problem.
Proceedings ArticleDOI
Learning to impersonate
Moni Naor,Guy N. Rothblum +1 more
TL;DR: If one-way functions do not exist, then an efficient Eve can learn to impersonate any efficient Bob nearly as well as an unbounded Eve, and tightly bound the number of observations Eve makes in terms of the secret's entropy.
Posted Content
On the Robustness of Information-Theoretic Privacy Measures and Mechanisms
TL;DR: It is proved that the optimal privacy mechanisms for the empirical distribution approach the corresponding mechanism for the true distribution as the sample size of the privacy mechanism increases, thereby establishing the statistical consistency of the optimalprivacy mechanisms.
Proceedings Article
PAC Confidence Sets for Deep Neural Networks via Calibrated Prediction
TL;DR: This work proposes an algorithm combining calibrated prediction and generalization bounds from learning theory to construct confidence sets for deep neural networks with PAC guarantees---i.e., the confidence set for a given input contains the true label with high probability.
Book ChapterDOI
Consistent Identification in the Limit of Rigid Grammars from Strings Is NP-hard
TL;DR: It is shown that the learning functions for these learnable classes are all NP-hard.