Coincidence of the $\sigma$-Algebras in a Scheme of Partially Observable Diffusion Processes

1988 ◽  
Vol 32 (3) ◽  
pp. 527-531
Author(s):  
T. A. Torondzhadze
2002 ◽  
Vol 14 (7) ◽  
pp. 1507-1544 ◽  
Author(s):  
Javier R. Movellan ◽  
Paul Mineiro ◽  
R. J. Williams

We present a Monte Carlo approach for training partially observable diffusion processes. We apply the approach to diffusion networks, a stochastic version of continuous recurrent neural networks. The approach is aimed at learning probability distributions of continuous paths, not just expected values. Interestingly, the relevant activation statistics used by the learning rule presented here are inner products in the Hilbert space of square integrable functions. These inner products can be computed using Hebbian operations and do not require backpropagation of error signals. Moreover, standard kernel methods could potentially be applied to compute such inner products. We propose that the main reason that recurrent neural networks have not worked well in engineering applications (e.g., speech recognition) is that they implicitly rely on a very simplistic likelihood model. The diffusion network approach proposed here is much richer and may open new avenues for applications of recurrent neural networks. We present some analysis and simulations to support this view. Very encouraging results were obtained on a visual speech recognition task in which neural networks outperformed hidden Markov models.


1976 ◽  
Vol 32 ◽  
pp. 109-116 ◽  
Author(s):  
S. Vauclair

This paper gives the first results of a work in progress, in collaboration with G. Michaud and G. Vauclair. It is a first attempt to compute the effects of meridional circulation and turbulence on diffusion processes in stellar envelopes. Computations have been made for a 2 Mʘstar, which lies in the Am - δ Scuti region of the HR diagram.Let us recall that in Am stars diffusion cannot occur between the two outer convection zones, contrary to what was assumed by Watson (1970, 1971) and Smith (1971), since they are linked by overshooting (Latour, 1972; Toomre et al., 1975). But diffusion may occur at the bottom of the second convection zone. According to Vauclair et al. (1974), the second convection zone, due to He II ionization, disappears after a time equal to the helium diffusion time, and then diffusion may happen at the bottom of the first convection zone, so that the arguments by Watson and Smith are preserved.


Author(s):  
Ari Arapostathis ◽  
Vivek S. Borkar ◽  
Mrinal K. Ghosh

1980 ◽  
Vol 41 (C6) ◽  
pp. C6-28-C6-31 ◽  
Author(s):  
R. Messer ◽  
H. Birli ◽  
K. Differt

2020 ◽  
pp. 130-135
Author(s):  
D.N. Korotaev ◽  
K.N. Poleshchenko ◽  
E.N. Eremin ◽  
E.E. Tarasov

The wear resistance and wear characteristics of cluster-gradient architecture (CGA) nanostructured topocomposites are studied. The specifics of tribocontact interaction under microcutting conditions is considered. The reasons for retention of high wear resistance of this class of nanostructured topocomposites are studied. The mechanisms of energy dissipation from the tribocontact zone, due to the nanogeometry and the structural-phase structure of CGA topocomposites are analyzed. The role of triboactivated deformation and diffusion processes in providing increased wear resistance of carbide-based topocomposites is shown. They are tested under the conditions of blade processing of heat-resistant titanium alloy.


Sign in / Sign up

Export Citation Format

Share Document