For semi-supervised techniques to be applied safely in practice we at least want methods to outperform their supervised counterparts.We study this question for classification using the well-known quadratic surrogate loss function. Unlike other approaches to semisupervised learning, the procedure proposed in this work does not rely on assumptions that are not intrinsic to the classifier at hand....

In the estimation of mean matrix in a multivariate normal distribution, generalized Bayes estimators with closed forms are provided, and sufficient conditions for their minimaxity derived relative to both scalar quadratic loss functions. The covariance also given forms, dominance properties discussed Stein function.

the paper describes an investigation about the application of the new multiple-impedance discontinuities model for optimizing profile diffuser barriers. the new multiple-impedance discontinuities model is much faster than the numerical method, thus it is used in an optimization process. the a-weighted insertion loss was used for the traffic noise spectrum. the result of optimization, which is d...

É proibida a reprodução parcial ou integral do conteúdo deste documento por qualquer meio de distribuição, digital ou im-presso, sem a expressa autorização do Insper ou de seu autor. A reprodução para fins didáticos é permitida observando-sea citação completa do documento Abstract This paper examines a set of inflation-targeting countries and presents evidence on the attitude of policymakers to...

Ridge regression, being based on the minimization of a quadratic loss function, is sensitive to outliers. Current proposals for robust ridge regression estimators are sensitive to “bad leverage observations”, cannot be employed when the number of predictors p is larger than the number of observations n; and have a low robustness when the ratio p=n is large. In this paper a ridge regression esti...

A very significant result in modern control theory is that, for a linear. finite-dimensional, dynamical system. the state feedback law derived from a quadratic-loss-function niinimisation problem is linear. The paper applies the results of this optimal control theory to a class of problems in which the feedback law is realised by a linear dynamical system. The quadratic loss function of interes...

In this paper we prove global convergence for first and second-order stationary points of a class of derivative-free trust-region methods for unconstrained optimization. These methods are based on the sequential minimization of quadratic (or linear) models built from evaluating the objective function at sample sets. The derivative-free models are required to satisfy Taylor-type bounds but, apar...

This paper presents allocation of power losses to consumers connected to radial distribution network before and after network reconfiguration in a deregulated environment. Loss allocation is made in a quadratic way, which is based on identifying the real and imaginary parts of current in each branch and losses are allocated to consumers. Comparison of loss allocation after multi-objective appro...

In this paper, the well-known competitive clustering algorithm (CA) is revisited and reformulated from a point of view of entropy minimization. That is, the second term of the objective function in CA can be seen as quadratic or second-order entropy. Along this novel explanation, two generalized competitive clustering algorithms inspired by Renyi entropy and Shannon entropy, i.e. RECA and SECA,...