Retrospective analyses of Pacific Basin climate records highlight the existence of a pan-Pacific interdecadal climate oscillation. We find strong evidence for coherent patterns of interdecadal variability in Pacific winds, sea level pressures, and upper ocean temperatures. Collectively, the ocean-atmosphere pattern of variability has been labeled the “Pacific Decadal Oscillation”, or PDO. An in...

AbsIracf-In this paper, the generalization ability 01 empirical risk minimization algorithms is investigated in the cootext of dislribution-lm prohably appmxjmalely comecl (PAC) learning. We identily a class of empirical risk minimization algorithms that are PAC, and show that the generic version of the support vector regression method belongs lo the class lor any given Mercer kernel. Moreover,...

A commonly used approach to multiclass classification is to replace the 0− 1 loss with a convex surrogate so as to make empirical risk minimization computationally tractable. Previous work has uncovered sufficient and necessary conditions for the consistency of the resulting procedures. In this paper, we strengthen these results by showing how the 0− 1 excess loss of a predictor can be upper bo...

Multiclass learning is an area of growing practical relevance, for which the currently available theory is still far from providing satisfactory understanding. We study the learnability of multiclass prediction, and derive upper and lower bounds on the sample complexity of multiclass hypothesis classes in different learning models: batch/online, realizable/unrealizable, full information/bandit ...

We develop an approach to risk minimization and stochastic optimization that pro-1vides a convex surrogate for variance, allowing near-optimal and computationally2efficient trading between approximation and estimation error. Our approach builds3off of techniques for distributionally robust optimization and Owen’s empirical4likelihood, and we provide a number of f...

für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution 4.0 International License. Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschung in Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht: Creative Commons Namen...

We consider a multiple sensor system such that for each sensor the outputs are related to the actual feature values according to a certain probability distribution. We present an overview of informational and computational aspects of a fuser that is required to combine the sensor outputs to more accurately predict the feature, when the sensor distributions are unknown but iid measurements are g...

We introduce a novel model for spatially varying variational data fusion, driven by point-wise confidence values. The proposed model allows for the joint estimation of the data and the confidence values based on the spatial coherence of the data. We discuss the main properties of the introduced model as well as suitable algorithms for estimating the solution of the corresponding biconvex minimi...

EFFICIENCY OF CLASSIFICATION METHODS BASED ON EMPIRICAL RISK MINIMIZATION V. I. Norkin a and M. A. Keyzer b UDC 519:234:24:85 A binary classification problem is reduced to the minimization of convex regularized empirical risk functionals in a reproducing kernel Hilbert space. The solution is searched for in the form of a finite linear combination of kernel support functions (Vapnik’s support ve...

In this lecture, we will introduce our second paradigm for document retrieval: probabilistic retrieval. We will focus on Roberston and Spärck Jones’ 1976 version, presented in the paper Relevance Weighting of Search Terms. This was an influential paper that was published when the Vector Space Model was first being developed — it is important to keep in mind the differences and similarities betw...