Recent work introduced loss functions which measure the error of a prediction based on multiple simultaneous observations or outcomes. In this paper, we explore the theoretical and practical questions that arise when using such multi-observation losses for regression on data sets of (x, y) pairs. When a loss depends on only one observation, the average empirical loss decomposes by applying the ...

It has been recently shown that, under the margin (or low noise) assumption, there exist classifiers attaining fast rates of convergence of the excess Bayes risk, i.e., the rates faster than n−1/2. The works on this subject suggested the following two conjectures: (i) the best achievable fast rate is of the order n−1, and (ii) the plug-in classifiers generally converge slower than the classifie...

We propose a novel framework for the differentially private ERM, input perturbation. Existing differentially private ERM implicitly assumed that the data contributors submit their private data to a database expecting that the database invokes a differentially private mechanism for publication of the learned model. In input perturbation, each data contributor independently randomizes her/his dat...

Improving the operational efficiency of processes is an important goal of business process management (BPM). There exist many proposals with regard to process improvement patterns (PIPs) as practices that aim at supporting this task. Nevertheless, there is still a gap with respect to validating PIPs in terms of their actual business value for a specific organization. Based on empirical research...

Despite widespread assertions that enthusiasm is an important quality of effective teaching, empirical research on the effect of enthusiasm on learning and memory is mixed and largely inconclusive. To help resolve these inconsistencies, we conducted a carefully-controlled laboratory experiment, investigating whether enthusiastic instructions for a memory task would improve recall accuracy. Scri...

In this paper, time series prediction is considered as a problem of missing values. A new method for the determination of the missing time series values is presented. The new method is based on two projection methods: a nonlinear one (Self-Organized Maps) and a linear one (Empirical Orthogonal Functions). The presented global methodology combines the advantages of both methods to get accurate c...

Learnability has always been one of the most central problems in learning theory. Most previous studies on this issue were based on the assumption that the samples are drawn independently and identically according to an underlying (unknown) distribution. The i.i.d. assumption, however, does not hold in many real applications. In this paper, we study the learnability of problems where the sample...

The Ruedenberg type formula relating the total molecular energy to the sum of orbital energies is examined by using SCC-MO and ab initio D Z M O eigenvalues. Comparison with rigorous ab initio D Z energy expectation values indicates that Ruedenberg 's formula in its original form can not provide semiquantitative information on molecular energetics. Much more promissing in this respect is the el...

Recall that: L(w) = 1 n E‖Xw − Y ‖ = 1 n E‖Xw − E[Y ]‖ + σ Define our “empirical loss” as: L̂(w) = 1 n ‖Xw − Y ‖ which has no expectation over Y . Note that for a fixed w E[L̂(w)] = L(w) e.g. the empirical loss is an unbiased estimate of the true loss. Suppose we knew the support size q. One algorithm is to simply find the estimator which minimizes the empirical loss and has support only on q coo...