N. Alon and J. Spencer, The Probabilistic Method. New York: Wiley, 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and the value of each weight is restricted to either 1 or Gamma1. Local search is applied to every new offspring, but only the weights connecting to the network s outputs will be changed by the local search. To a certain extent, the local search method in [25] is similar to Hebbian learning [29]. For each weight connecting to an output node, there is an associated variable d that is initialized to zero before applying the local search. When the training patterns are fed to the network one at a time during the application of local search, d will be increased if the activations of the two ....

D. O. Hebb. The Organization of Behavior. New York: Wiley, 1949.

....all . Our measure of the complexity of an encoder or decoder function is the minimum number of computation steps (or gates ) in a logic circuit that computes the function. We use a standard definition of a Boolean circuit containing ary gates that belong to a specified basis, as in, for example, [29], 30] If the circuit computes a function of Boolean inputs , then starting with the inputs and the Boolean constants and , the circuit can compute new functions by applying a function in the basis to any two previously computed functions. 1) Let (the basis) be a set of ary Boolean functions ....

J. E. Savage, The Complexity of Computing. New York: Wiley, 1976.

....This section illustrates the main paradigm of the approach for logic decomposition. First, some background on tree height reduction is presented. Next, a step in the recursive progress of the main algorithm is described with an example. A. Tree Height Reduction: An Example Tree height reduction [16] was originally proposed in the scope of optimizing compilers for the generation of code in multiprocessor systems. Fig. 2. illustrates an example. The tree in Fig. 2(a) represents a factored form for the Boolean expression (1) If we assume zero arrival time for all inputs and unit area and unit ....

D. Kuck, The Structure of Computers and Computation. New York: Wiley, 1978.

....to help students more easily view their data transformation during program execution. However, it is only designed to work for Pascal programs and does not handle sophisticated data structures (such as linked lists or records) The use of animation to show program execution is not a new idea [3,7,9]. Algorithm animation (e.g. XTANGO [13] BALSA [4] and program visualization [7] has been developed with the idea of incorporating visualization into the learning process. The instructor programs the algorithm animation. The student just runs the program, observing the animation of the algorithm ....

....just runs the program, observing the animation of the algorithm when using different inputs. Thus, algorithm animation packages generally do not help students in the visualization of their own programs. One of the most successful educational simulation software packages has been Karel, The Robot [9]. Karel was used to prepare students for success in learning Pascal. Students were able to view the entire world (environment) of Karel, and watch how the world changed as their programs executed. Over the last decade, curricula for CS1 have made a transition to C and then to object oriented ....

Pattis, R., Karel the Robot. New York: Wiley, 1981.

....Visualization To Teach Novices Recursion Wanda Dann Computer Science Dept. Ithaca College Ithaca, NY 14850 t 607 274 3602 wpdan n Ithaca.edu Stephen Cooper Computer Science Dept. Saint Joseph s University Philadelphia, PA 19131 1 610 660 1561 scooper sju.edu Randy Pausch Computer Science Dept. Camegie Mellon University Pittsburgh, PA 15213 1 412268 3579 pausch cs.cmu.edu ABSTRACT ....

....Program visualization tools differ from algorithm animation tools in that program visualization direcQy relates individual lines of a student s own program code to the animation. Some general purpose program visualization tools have been designed for use by beginners (e.g. Karel, The Robot [14] Karel was originally used to prepare students for success in learning Pascal, and has undergone several update, the latest being Karel [1] KareN can be used to introduce recursion by writing taii recursive functions for the robot. Similarly, LOGO [13] can be used to demonstrate some ....

Pattis, R, Karel the Robot. New York: Wiley, 1981.

....[9] for a review) Nonlineaxities can be modeled implidtly (such as with differential equations) or explicitly, and our interest lies in the latter. A broad class of nonllneax systems have polynomial functional representations, for which the Volterra and the Wiener representations are two examples [11]. An Lth order discrete time Volterra nonlinear system relates the output x(t) and the input w(t) via multi dimensional convolutions, which are straightforward gen eralizations of the 1 D convolution: x uO In (1) h0 is the dc term, t) is the additive noise assumed to be zero mean ....

.... different form, where the functionsis are made orthogonal to each other through a Gram Schmidt orthogonMization procedure, with respect to zero mean white Gaussian input w(t) The orthogonal ity property facilitates closed form solutions and sepaxates kernel estimation through cross correlations [11]. However, the model becomes increasingly complex with larger L and the theory does not apply to non Gaussiaa inputs. In Volterra nonlinear system identification with stochas tic inputs, a lot of attention has been given to Gaussian inputs because their higher order moments are tractable. In this ....

M. Sezen The Volierra and Wiener Theories of Nonlineg Ssiems. New York: Wiley 1980.

....functions. Several density estimation can be used, including parametric methods based, e.g. on a Gaussian model, and non parametric kernel methods. In the simulations presented in the sequel, we chose to use a Gaussian mixture model together with the EM algorithm as an estimation technique [12]. As remarked by Bastire [8] there is no general technique for evaluating the discounting coe cients q in the separable method. In this paper, we propose to use the same approach as used by Denoeux [4] for the DB method, i.e. minimizing the following error criterion: E( n X i=1 Q X ....

.... not shown here for lack of space) In these gures, mixture component centers and prototypes are represented as asterisks ( For the LB methods, likelihood functions were estimated using a Gaussian mixture model with k = 2 modes per class, and the parameters were estimated by the EM algorithm [12]. For the DB method, we chose by analogy two prototypes per class whose locations were initialized using the c means algorithm. Concerning the separable method, parameters q were xed at the following values: 1 = 0:4, 2 = 3 = 0:9. The value of the rejection cost 0 was set at 0:4. The ....

G. J. M. Lachlan and T. Krishnan, The EM Algorithm and Extensions. New-York: Wiley, 1997.

.... to [Dreyfus 79] and Chapter 1 of [Shavlik and Dietterich 90] for good historical reviews) NNs made their first real appearance with Rosenblatt s perceptron in the 1950 s (see [Rosenblatt 58, Rosenblatt 62, Minsky and Papert 67] as a result of earlier studies on neurons [McCulloch and Pitts 43, Hebb 49] Researchers in AI ML and NNs had the ambition, hope, and enthusiasm characteristic of most newlyborn fields of research. However, the optimism of the early days was soon replaced by the hard reality of the extreme complexity of the task at hand. Both fields met with violent criticism, both of a ....

Hebb, D.O. The Organization of Behavior. New York: Wiley, 1949.

....the experimenters conjecture that different attentional states were associated with different interaction structures. 1 INTRODUCTION Simultaneous activation of two or more neurons is, as Aertsen and Grn say, the unitary event for information processing in the brain [1] Since the time of Hebb s [2] fundamental theory it has been understood that complex information processing in the brain arises from the collective interaction of groups of neurons. Experimental advances in the last decade enable the direct study of firing patterns among the spiking events of groups of neurons. Abeles and his ....

....or rejected, with the probability of acceptance given by: min 1, p(x 1 ,K.x N ) p(x 1 ,K.x N ) 15) 5. APPLICATION OF THE METHODS There is growing consensus that processing in the brain is organized in functional groups of neurons. Following Hebb [2], these groups are commonly referred to as cell assemblies. One operational definition for the cell assembly has been particularly influential: nearsimultaneity or some other specific timing relation in the firing of the participating neurons. Such temporal coherence is at least in principle ....

Hebb, D. The Organization of Behavior. New York: Wiley, 1949.

....that this processing replaces the silver grains in the red, green, and blue layers with cyan, magenta, and yellow dyes in accordance with the principles of subtractive color reproduction, which were be discussed in Section III A2. A more complete description of color photography can be found in [130], and simplified mathematical models for the process are described in [52, pp. 335 339] As an aside, one may note that prior to the invention of spectrophotometers and spectroradiometers, two techniques were developed to record the spectral information of entire images on (monochromatic) film. ....

....a standing wave pattern by using a mercury coating on the rear of the film as a mirror. Both methods required extremely fine grain film in order to achieve the high resolution required and long exposure times to compensate for the low energy at each spectral wavelength. The reader is referred to [130] for a slightly more detailed account of these techniques. 3) Colorimeters, Cameras, and Scanners: Colorimeters, digital color cameras, and color scanners are color recording devices that operate on very similar principles and differ primarily only in their intended use. All these devices record ....

R. W. G. Hunt, The Reproduction of Color, 3rd ed. New York: Wiley, 1975.

....or social choice function, is some function , such that represents the preferences of the group. In Harsanyi s formulation of social choice, each individual s preferences are represented by a von Neumann Morgenstern cardinal utility function that obeys the standard axioms of Bayesian rationality [11]. In particular, is defined to be the weighted summation of the individual utilities (1) where individual agent; weight given to agent utility function in the overall equation; action under consideration. To be socially rational, an individual maximizes over the different alternatives. This ....

L. J. Savage, The Foundations of Statistics. New York: Wiley, 1954.

....in N(n; 2; z) tends to in nity. Based on the Chebychev s inequality, to prove that N(n; 2; z) contains t 3 modules with probability 1, we need to show that the variance of A t , the number of contained t 3 modules, is o(EfA t g) To do this, we need the following Lemma 3.3.6. Alon and Spencer [ALS92]) Given a random structure(e.g. a random CNF formula) let W be the set of substructures under consideration, A(w) be the set of substructures sharing some clauses with w 2 W . Let I w = 1 when w is in the random structure and 0 otherwise. If (1) elements of W are symmetric; 2) Ef P w2W ....

N.Alon and J.H.Spencer, The Probabilistic Method, New York: Wiley, 1992.

....under Knightian uncertainty are discussed. Keywords: Knightian uncertainty# belief functions# entropy# nonparametric methods, Prohorov metric, core diameter. 1 Introduction The immense success of the von Neumann Morgenstern Savage decision theory paradigm (see Neumann and Morgenstern (1947) and Savage (1954)) can be partly attributed to the essential reduction of all situations of uncertaintyto situations of risk with known probabilities on events. Henceforth, we shall refer to Knightian uncertainty (Knight (1921) or ambiguity to describe situations in which no objective probability measure on ....

Savage, L. (1954): The foundations of statistics. New York: Wiley.

....in the form outlined in Figs. 1 and 2. Therefore, throughout Sections III V of this work, we will consider a reduced form of this problem, one in which a process operates at a single setting. This reduced problem allows us to consider these methods in the context of traditional statistical theory [6], 7] We will later return to the case of a continuous response to multiple process settings in Section VI. In our reduced form of the nonuniformity estimation problem, we assume sites on each of the wafers that are measured and that the thickness of the resulting film is given SMITH et al. ....

....the expected value of the SRS and MRS estimators, while Section IV outlines the determination of the variance of these estimators. In variancecomponents theory, the form given in (1) is used to understand these same components of variance which arise at the waferto wafer and within wafer levels [6], 7] These works outline how the within wafer variance is nested within the wafer towafer variance. They also outline techniques for hypothesis testing, estimating confidence intervals, and for generating unbiased point estimates of the components of variance (by equating the moments determined ....

H. Scheff e, The Analysis of Variance. New York: Wiley, 1959.

No context found.

N. Alon and J. Spencer, The Probabilistic Method. New York: Wiley, 1992.

No context found.

Hebb D. O. The Organization of Behavior. New York :Wiley , 1949

No context found.

R. W. G. Hunt, The Reproduction of Color, 3rd ed. New York: Wiley, 1975.

No context found.

D. O. Hebb, The Organization of Behavior. New York: Wiley, 1949.

No context found.

D.O. Hebb, The Organization of Behavior, New York: Wiley, 1949.

No context found.

E. Klinger, The structure and function of fantasy. New York: Wiley, 1971.

No context found.

Hebb, D. The Organization of Behavior. New York: Wiley, 1949.

No context found.

J E Savage, The Complexity of Computing, New York: Wiley, 1976.

No context found.

D J Kuck, The Structure of Computers and Computations, New York: Wiley, 1978.

No context found.

: The Foundations of Statistics. New York: Wiley.

No context found.

R. W. Boyd, Radiometry and the Detection of Optical Radiation. New York: Wiley, 1983.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC