Although great progresses have been made in automatic speech recognition (ASR), significant performance degradation is still observed when recognizing multi-talker mixed speech. In this paper, we propose and evaluate several architectures to address this problem under the assumption that only a single channel of mixed signal is available. Our technique extends permutation invariant training (PI...

We find that Ce(Ni(0.25)In(1.75)) crystallizes in the hexagonal AlB(2)-type structure with lattice parameters a = 0.4850(5) nm and c = 0.3908(5) nm. Magnetic susceptibility, electrical resistivity and low-temperature specific heat data reveal that the bulk phase transition at 3.7 ± 0.2 K in Ce(Ni(0.25)In(1.75)) is to an antiferromagnetic state. The magnetic contribution to the resistivity ρ(mag...

Degrees of belief are formed using observed ev­ idence and stat ist ical background in format ion. In this paper we examine the process of how pr ior degrees of belief derived f rom the evidence are combined w i th stat ist ical data to form more specific degrees of belief. A stat ist ical model for this process then is shown to vindicate the cross-entropy min imizat ion principle as a rule for...

We show that extensive thermostatistics based on Rényi entropy and Kolmogorov–Nagumo averages can be expressed in terms of Tsallis nonextensive thermostatistics. We use this correspondence to generalize thermostatistics to a large class of Kolmogorov–Nagumo means and suitably adapted definitions of entropy. As an application, we reanalyze linguistic data discussed in a paper by Montemurro.  20...

Based on the intuitionistic knowledge content characteristic of information gain, the concepts of combination entropy CE(A) and combination granulation CG(A) in incomplete information system are introduced, their some properties are given. Furthermore, the relationship between combination entropy and combination granulation is established. These concepts and properties are all special instances...

Conditioning is the generally agreed-upon method for updating probability distributions when one learns that an event is certainly true. But it has been argued that we need other rules, in particular the rule of cross-entropy minimization, to handle updates that involve uncertain information. In this paper we re-examine such a case: van Fraassen’s Judy Benjamin problem [1987], which in essence ...

We propose a moving edge extraction method using the concept of entropy and cross-entropy, in which the crossentropy concept is applied to dynamic scene analysis. The cross-entropy concept provides enhancement of detection for the dynamically changed area. We combine the results of cross-entropy in the difference picture (DP) with those of entropy in the current frame so that we can effectively...

A logic is defined that allows to express in­ formation about statistical probabilities and about degrees of belief in specific proposi­ tions. By interpreting the two types of proba­ bilities in one common probability space, the semantics given are well suited to model the influence of statistical information on the for­ mation of subjective beliefs. Cross entropy minimization is a key element...

General geodesic equations of the motion of spinning systems around the (3+1)-dimensional and (2+1)-dimensional rotating anti-de Sitter black holes have been obtained. Based upon these equations, we derived the entropy bound for a rotating system from Kerr-Anti de Sitter black holes and BTZ black holes, respectively. Our result coincides with that of Hod’s derived from Kerr black hole, which sh...

One of the basic problems in data analysis lies in choosing the optimal rescaling (change of coordinate system) to study properties of a given data-set Y . The classical Mahalanobis approach has its basis in the classical normalization/rescaling formula Y 3 y → Σ Y · (y − mY ), where mY denotes the mean of Y and ΣY the covariance matrix . Based on the cross-entropy we generalize this approach a...