Optimal Hoeffding-like inequalities under a symmetry assumption

Extensions of the Hoeffding-Azuma inequalities

In this paper we give extensions of the Hoeffding-Azuma inequalities for weighted sums of uniformly bounded martingale differences. Our results improve previous results of Antonov (1979).

Optimal rates of aggregation in classification under low noise assumption

Let (X ,A) be a measurable space. We consider a random variable (X,Y ) on X × {−1,1} with probability distribution denoted by π . Denote by P the marginal of π on X and by η(x) def = P(Y = 1|X = x) the conditional probability function of Y = 1, knowing that X = x. We have n i.i.d. observations of the couple (X,Y ) denoted by Dn = ((Xi, Yi))i=1,...,n. The aim is to predict the output label Y for...

SERIE B - INFORMATIK Algorithmic Chernoff-Hoeffding Inequalities in Integer Programming

Hoeffding Inequalities for Join Selectivity Estimation and Online Aggregation

We extend Hoe ding s inequalities for simple averages of random variables to the case of cross product averages We also survey some new and existing Hoe ding inequalities for estimators of the mean variance and standard deviation of a subpopulation These results are applicable to two problems in object relational database management systems xed precision estimation of the selectivity of a join ...

Optimal Classifier’s Independence Assumption 1 A Test of the Optimal Classifier’s Independence Assumption in Perceptual Categorization

Observers completed perceptual categorization tasks that included separate base-rate/payoff manipulations, corresponding simultaneous base-rate/payoff manipulations, and conflicting simultaneous base-rate/payoff manipulations. Performance was (a) closer to optimal for 2:1 than 3:1 base-rate/payoff ratios, and when base-rates, as opposed to payoffs were manipulated, and (b) was more in line with...