Showing papers on "Generalization published in 1984"

Scalar timing in memory.

Abstract: A recent report of ours’ proposed an information-processing account of temporal generalization. The account posited a clock process, which was the basic time measurement device, and working and reference memory for storing the output of the clock either temporarily or relatively permanently. Records of time intervals in working and reference memory were then compared using a binary decision process, which dictated responding or not responding. The analysis concentrated on a relativistic Weber’s law property of the data from temporal generalization, and the constraints this property imposed on sources of variance in the information-processing stages. Our purpose here is to summarize that work and generalize the model in two ways: First we consider several sources of variance operating simultaneously. The original analysis demonstrated that if only one source of variance is present, it must be a scalar source, that is, it must result in a variable memory for which variance increases with the square of the mean.’ In the generalized account proposed here, we will develop the conclusion that scalar sources dominate in some time ranges, while other sources may dominate in others. These ideas are then applied to two additional timing tasks with different characteristics.

Evolutionary stable strategies and game dynamics for n-person games

Abstract: This note contains a generalization of the definition of an evolutionary stable strategy and of the corresponding game dynamics from 2-person to n-person games. This broader framework also allows modelling of several interacting populations or of populations containing different “types” of individuals, for example males and females.

A class of direct methods for linear systems

Abstract: A class of methods of direct type for solving determined or underdetermined, full rank or deficient rank linear systems is presented and theoretically analyzed. The class can be considered as a generalization of the methods of Brent and Brown as restricted to linear systems and implicitly contains orthogonal,LU andLL T factorization methods.

On the fundamentals of fuzzy sets.

Abstract: A considerable amount of research has been done on the notions of pseudo complement, intersection and union of fuzzy sets [1], [4], [11]. Most of this work consists of generalizations or alternatives of the basic concepts introduced by L. A. Zadeh in his famous paper [13]: generalization of the unit interval to arbitrary complete and completely distributive lattices or to Boolean algebras [2]; alternatives to union and intersection using the concept of t-norms [3], [10]; alternative complements in the unit interval in [4], [11], a.s.o. Although it is usually accepted that I := [0,1] is a natural generalization of {0,1} and that a --> 1 - a is a natural generalization of the Boolean complement on {0,1}, we do however not find canonical and mathematical justification for this fact, which nevertheless lies at the heart of the definition of fuzzy sets. It is the purpose of this note to present a canonical way of obtaining I and L. A. Zadeh's pseudocomplement. Moreover, if we consistently use this canonical machine we shall see that also other set-concepts can, and maybe should have been extended to fuzzy sets. Further we also give a possible generalization involving a choice of arbitrary t-norms.

Concept Learning in a Rich Input Domain: Generalization-Based Memory

Abstract: AutomatIc concept learning from large amounts of complex input data IS an interestmg and difficult process. In this paper we discuss how the use of a permanent, generalization-based, memory can serve as an important tool In developing programs that learn in rich input domains. The use of GeneralizationBased ~emory (GBM) allows programs to determine what concepts to learn, as well as definitlOns of the concepts. We present in this paper a characterization of our research, descnbe our use of Generalization-Based Memory in two programs under development at Columbia, UNIMEM and RESEARCHER, and describe how they perform concept evaluation and generalizatIOn of complex structural descriptions, problems tYPical of those we are concerned wIth. Key Terms: Learmng, automatic concept formation, generalization, GeneralizatIon-Based Memory, intelligent InformatlOn systems, artificial intelligence, cognItIve SCIence

Lossless inverse scattering, digital filters, and estimation theory

Abstract: Methods to construct rational solutions of the lossless inverse scattering (LIS) problem for one-port passive digital systems are described. The first method is recursive and can be viewed as a generalization of the celebrated Schur algorithm. The second method is global and leads to a parametrization of what we call fundamental solutions from which all LIS solutions may be constructed. Quite a few classical problems in estimation theory and network theory may be viewed as special cases of the LIS problem. With each fundamental solution there is a solution of a corresponding estimation problem leading to a prediction and a modeling filter for a given stochastic process.

Constraint-based generalization: learning game-playing plans from single examples

Abstract: Constraint-based Generalization is a technique for deducing generalizations from a single example. We show how this technique can be used for learning tactical combinations in games and discuss an implementation which learns forced wins in tic-tac-toe, go-moku, and chess.

Complex space-time ray method and “quasiphotons”

Abstract: In the paper formal asymptotic solutions of the wave equation are constructed which represent concentrated wave packets propagating along rays. The construction technique is a time generalization of the frequently used spatial complex ray method. Analogously to the method of complex germ of V. P. Maslov, it is based on considering complex solutions of the Hamilton-Jacobi equation.

Corrigendum to our paper “the ellipsoid method and its consequences in combinatorial optimization”

Abstract: An error in the proof, and in the statement of a generalization, of the result that submodular setfunctions can be minimized over the subsets with odd cardinality is corrected.

Evaluation of Electrochemical Methods in Solid State Research and Their Generalization for Defects with Variable Charges

Abstract: The evaluation of instationary and stationary electrochemical techniques (d.c. and a.c. ; electrochemical and chemical polarization ; blocking and non-blocking electrodes ; two or multipoint arrangement) which allow the determination of transport quantities are discussed for mixed conductors, including the limiting cases of purely ionic or purely electronic conductance. Neglecting Onsager-coupling the evaluation formulae (instationary and stationary polarization, Wagner-Hebb-measurements, impedance measurements, concentration cell experiments, permeation technique etc.) are derived under the general aspect of solid state diffusion and system theory for a very general scheme of defect types and are generalized in so far as all ionic defects may change their charges by reacting with electronic defects. Because of the differences of the diffusion coefficients for the differently ionized defects and because of the coupling between electronic and ionic species new effects occur manifesting themselves in additional terms in the evaluation formulae. For the case ofconstant charges an equivalent circuit is given that describes all experiments under investigation.

An inventory model with finite‐range stochastic lead times

Abstract: This article considers an inventory model with constant demand and stochastic lead times distributed over a finite range. A generalization of the EOQ formula with backorders is derived and ranges for the decision variables are obtained. The results are illustrated with the case of uniformly distributed lead time.

An Introduction to Multilinear Formula Score Theory.

Abstract: : Formula score theory associates each multiple choice test with a linear operator and expresses all of the real functions of item response theory as linear combinations of the operator's eigenfunctions. Hard measurement problems can then often be reformulated as easier, standard mathematical problems. For example, the problem of estimating ability distributions from sequences of item responses can be reformulated as maximizing a convex index of goodness of fit defined on a convex set. A major simplification of several theoretical problems has been obtained because the linear mathematics used by the theory has a well-developed generalization to problems involving many variables. For example, a battery of tests measuring several related variables and one test measuring one trait can be analyzed with essentially the same theory. An elementary outline of the basic theory is presented along with a discussion of several illustrative applications. (Author)

Validity generalization: Some critical remarks on the Schmidt-Hunter procedure*

Abstract: This article presents some critical comments on the validity generalization procedure which has been presented by Schmidt, Hunter and others. They have put forward a method for testing the hypothesis that the variance in validity coefficients across situations for job-test combinations is due to what they con­sider to be statistical artifacts. The Schmidt-Hunter approach is criticized on the following points: the compilation of validity data, the use of criterion measures, and the test of the hypothesis of no situational specificity. Further, the relation between the concepts ‘ situational specificity ’ and ‘ validity generalization ’ is con­sidered. In addition, it is noted that Schmidt, Hunter and others have defined the concept ‘ situation ’ in a different way than classical writers. It is concluded that the Schmidt-Hunter approach to validity generalization shows fundamental shortcomings. As a consequence their far-reaching conclusions for the practice of personnel selection should be considered premature.

An algorithmic approach to non-homogeneous semi-markov processes

Abstract: De Dominicis 1979 suggested a generalization of the notion of semi-Markov process by introducing a suitable definition of the transition kernel of the process and proved a theorm on the existence and uniqueness of the solution of integral equation describing the process evolution. In this paper a quite general algorithm useful for the resolution of the evolution equation is performed; in addition an application to an actuarial problem is presented.

Constraint limited generalization: acquiring procedures from examples

Abstract: Generalization is an essential part of any system that can acquire knowledge from examples. I argue that generalization must be limited by a variety of constraints in order to be useful. This paper gives three principles on how generalization processes should be constrained. It also describes a system for acquiring procedures from examples which is based on these principles and is used to illustrate them.

Application of the “generalization map” to a self‐control intervention with school‐aged children

Abstract: The assessment of generalization has become a priority of applied behavior analysis. This study provided a thorough assessment of the generality of a comprehensive self-control intervention. This intervention incorporated a number of self-management skills and was designed to increase the math performance of an underachieving student in a regular elementary school classroom. All possible classes of generalization as outlined by Drabman, Hammer, and Rosenbaum (1979) were assessed. An ABAB design with follow-up was used to determine the effectiveness of the intervention for the treated student's math performance in the school setting as well as the degree of generalization across the following untreated dimensions: behavior (disruptiveness); setting (home); subject (classmate); and time period (follow-up). The effective intervention produced: subject, behavior, subject-behavior, setting, subject-setting, behavior-setting, subject-behavior-setting, time, subject-time, setting-time, subject-setting-time, and subject-behavior-setting-time generalization. Generalization was not obtained for behavior-time, subject-behavior-time, and behavior-setting-time generalization. Features of this intervention which may have promoted generalization are discussed.

First integrals for some nonlinear time‐dependent Hamiltonian systems

Abstract: An observation of a simple property of canonical transformations leads to a procedure for determining first integrals for classes of Hamiltonians. In illustrative examples the most general result presently known from other methods is recovered, a new result presented, and a generalization to more than one degree of freedom discussed.

The Effect of Key Behavior Distinctiveness on Generalization and Recall in Behavior Modeling Training

Abstract: The article focuses on a study which analyzed the effect of key behavior distinctiveness on generalization and recall in behavior modeling training. Subjects were asked to view the videotape chosen...