Results 301 to 310 of about 1,699,991 (339)

Bivariate probability-based anomaly detection

2014 International Conference on Behavioral, Economic, and Socio-Cultural Computing (BESC2014), 2014
Statistical techniques play a crucial role in anomaly detection. Although they usually are simple and can be trained unsupervised, they face three challenges: parametric techniques usually rely on the assumption that the data meet a special distribution; existing Histogram-based techniques only take account of individual attribute, which cannot capture
Hua Lou, Ye Zhu 0002
openaire   +1 more source

Probability detection mechanisms and motor learning

Experimental Brain Research, 2004
The automatic detection of patterns or regularities in the environment is central to certain forms of motor learning, which are largely procedural and implicit. The rules underlying the detection and use of probabilistic information in the perceptual-motor domain are largely unknown.
O V, Lungu, T, Wächter, T, Liu, D T, Willingham, J, Ashe   +4 more
openaire   +2 more sources

On the Probability of Detecting a Region by a Linear Search

Theory of Probability & Its Applications, 1964
Lying in a circle G is a convex rectifiable region K. If a cut L is made in the circle at random then the probability of cutting the region K is \[ P = \frac{{\frac{2}{\pi }LV(K) + W(K)}}{{\pi R^2 - L\sqrt {R^2 - \frac{{L^2 }}{4}} - R^2 \arcsin \frac{L}{{2R}}}}. \] Here, $V(K)$ and $W(K)$ are the linear and planar variations of K in the sense of A.
openaire   +2 more sources

Detection error probabilities in noncoherent channels

Proceedings. International Symposium on Information Theory, 2005. ISIT 2005., 2005
This paper presents performance bounds for block-based detection of PAM and QAM transmitted over non-coherent channels. Such channels give rise to inherent detection ambiguities since multiple possible transmit sequences can have the same likelihood. The number of ambiguous sequences is a function of the block length.
Ryan, D.J., Clarkson, I.V.L., Collings, I.B.   +2 more
openaire   +2 more sources

Probability of Detection and Reliability

2016
Ultrasonic nondestructive evaluation is often used to help guarantee safety and reliability of critical structures and systems. Thus, it is natural to leverage the use of ultrasonic models in the determination of inspection parameters such as probability of detection (POD) and in estimating the influence that inspections have on reliability.
openaire   +1 more source

Probability of Signal Detection in a Vigilance Task

Science, 1962
It is hypothesized that the probability of detecting a signal in a vigilance task depends upon its temporal location with respect to the preceding series of signals. Probability of detection should be at a maximum when the signal occurs after a temporal interval which is equivalent to the mean of the intervals between the preceding signals detected ...
openaire   +2 more sources

Computing the Probability of False Watermark Detection

2000
Several methods of watermark detection involve computing a vector from some input media, computing the normalized correlation between that vector and a predefined watermark vector, and comparing the result against a threshold. We show that, if the probability density function of vectors that arise from random, unwatermarked media is a zero-mean ...
Matthew L. Miller, Jeffrey A. Bloom
openaire   +1 more source

Outlier Detection with Arbitrary Probability Functions

2013
We consider the problem of unsupervised outlier detection in large collections of data objects when objects are modeled by means of arbitrary multidimensional probability density functions. Specifically, we present a novel definition of outlier in the context of uncertain data under the attribute level uncertainty model, according to which an uncertain
Angiulli F, FASSETTI, Fabio
openaire   +3 more sources

Detection probabilities for a rank statistic (Corresp.)

IEEE Transactions on Information Theory, 1971
Two statistics r = [x_{(n)} - x_{(n-1)}]/[x_{(n-1)} - x_{(1)}] and u = x_{(n)} , where x_{(i)} is the i th largest order statistic of a sample of size n , are studied to determine whether a preassigned false-alarm probability can be achieved in signal detection and to determine their probabilities of detection, of which graphs are presented.
openaire   +1 more source

The detection probabilities in simultaneous pursuit games

1994
One could expect that the difference in detection probabilities for optimal pursuit and search strategies will be small for small l (detection radius), but the above examples show that this is not always case. This suggests that the choosing of the payoff function must be done with special care by the applied mathematitian, because the intuitive ...
openaire   +1 more source