algorithms. It is shown that the essential condition for distribution-free learnability is finiteness of the Vapnik-Chervonenkis dimension, a simple combinatorial parameter of the class of concepts to be learned. Using this parameter, the complexity and closure properties of learnable classes are analyzed

manifold, beamforming is made robust with respect to array deformations, distortion of the wave front, pointing errors, etc ... so that neither array calibration nor physical modeling are necessary. Rather surprisingly, `blind beamformers' may outperform `informed beamformers' in a plausible

. Again we found that loopy belief propagation always converged, with the average number of iterations equal to 8.65. The protocol for the ALARM network experiments dif fered from the previous two in that the structure and parameters were fixed -only the observed evidence differed between experimental

We consider policies for CPU load balancing in networks of workstations. We address the question of whether preemptive migration (migrating active processes) is necessary, or whether remote execution (migrating processes only at the time of birth) is sufficient for load balancing. We show

of the principal points, pixels aspect ratio and focal lengths). This is important for two reasons. First, it is more realistic in applications where these parameters may vary according to the task (active vision). Second, the general case considered here, captures all the relevant information that is necessary

In a Bayesian mixture model it is not necessary a priori to limit the number of components to be finite. In this paper an infinite Gaussian mixture model is presented which neatly sidesteps the difficult problem of finding the "right" number of mixture components. Inference in the model

of the paper, that all parameters are indeed integers. We then give as our main lemma both necessary and sufficient conditions for a task system to be feasible on one processor. Although these conditions cannot, in general, be tested efficiently (unless P = NP), they do allow us to give efficient algorithms

To test the hypothesis of a di erence stationary time series against a trend stationary alternative, Levin and Lin (1993) and Im, Pesaran and Shin (1997) suggest bias adjusted t-statistics. Such corrections are necessary to account for the nonzero mean of the t-statistic in the case of an OLS

parameters (see Table 1 for their estimated values): . vocabulary size: number of concepts necessary to formalize all descriptions in the repository; . description complexity: average number of concepts for one description; . description heterogeneity: average number of semantic relations in a

in storage subsystem design require similar levels of detail. This report describes a suite of general-purpose techniques and algorithms for acquiring the necessary data from SCSI disks via the ANSI-standard interface. The accuracy of the extracted information is demonstrated by comparing the behavior of a