In this report, we address the problem of parameter learning for belief networks with fixed structure based on empirical observations. Both complete and incomplete (data) observations are included. Given complete data, we describe the simple problem of single parameter learning for intuition and then expand to belief networks under appropriate system decomposition. If the observations are incom...

Reading a book is also kind of better solution when you have no enough money or time to get your own adventure. This is one of the reasons we show the neural fuzzy control systems with structure and parameter learning as your friend in spending the time. For more representative collections, this book not only offers it's strategically book resource. It can be a good friend, really good friend w...

Crystal structure and defects have been shown to have a strong impact on III-V nanowire properties. Recently, it was demonstrated that the issue of random stacking and polytypism in semiconductor nanowires can often be controlled using accessible growth parameters such as temperature, diameter, and V/III ratio . In addition, it has been shown that crystal phase can be tuned selectively between ...

We present a robust and computationally efficient parameter estimation strategy for fluid-structure interaction problems. The method is based on a filtering algorithm restricted to the parameter space, known as the reduced-order unscented Kalman filter. It does not require any adjoint or tangent problems. In addition, it can easily be run in parallel, which is of great interest in fluid-structu...

We present an online learning approach for robustly combining unreliable observations from a pedestrian detector to estimate the rough 3D scene geometry from video sequences of a static camera. Our approach is based on an entropy modelling framework, which allows to simultaneously adapt the detector parameters, such that the expected information gain about the scene structure is maximised. As a...

where x = x±x and p = p+p refer to the usual light-cone components, p̄ = 1 2 (p+ p), ∆ = p− p, and t ≡ ∆. The “off-forwardness” (or skewedness) in Eqs. (1,2) is defined as ξ ≡ ∆ + p+ . From the point of view of parton physics in the infinite momentum frame, these OFPDs have the physical meaning of the amplitude for the process that a quark is taken out of the nucleon with longitudinal momentum f...

There exist a number of provably correct learning algorithms for Optimality Theory and closely related theories. These include Constraint Demotion (CD; Tesar 1995, et seq.), a family of algorithms for classic OT. For Harmonic Grammar (Legendre, Miyata and Smolensky 1990; Smolensky and Legendre 2006) and related theories (e.g. maximum entropy), there is Stochastic Gradient Ascent (SGA; Soderstro...

In this article, regulation of a distributed-parameter flexible beam is considered using variable structure control techniques. The proposed controller can stabilize the system exponentially and the converging speed can be set by the designer as desired. Different from existing variable structure controllers for flexible robots in the literature, the controller presented here is designed direct...

In many applications, such as those arising from the field of cellular networks, it is often desired to determine the interaction (graph) structure of a set of differential equations, using as data measured sensitivities. This note proposes an approach to this problem.

A fundamental challenge for any general theory of neural circuits is how to characterize the structure of the space of all possible circuits over a given model neuron. As a first step in this direction, this letter begins a systematic study of the global parameter space structure of continuous-time recurrent neural networks (CTRNNs), a class of neural models that is simple but dynamically unive...