By Topic

Systems Science and Cybernetics, IEEE Transactions on

Issue 3 • Date July 1969

Filter Results

Displaying Results 1 - 20 of 20
  • [Table of contents]

    Publication Year: 1969 , Page(s): c1
    Save to Project icon | Request Permissions | PDF file iconPDF (93 KB)  
    Freely Available from IEEE
  • IEEE Systems Science and Cybernetics Group

    Publication Year: 1969 , Page(s): c2
    Save to Project icon | Request Permissions | PDF file iconPDF (133 KB)  
    Freely Available from IEEE
  • Some New Approaches to Machine Learning

    Publication Year: 1969 , Page(s): 173 - 182
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2741 KB)  

    Five different but interrelated models of learning have been established within a complex computer program. These models incorporate mechanisms that optimize response patterns on algorithmic and heuristic bases; make abstractions at different levels; produce value judgements; recognize, modify, store, and retrieve geometrical patterns; and exhibit, in general, many aspects of intelligent behavior. Both the teacher and the learner are simulated in the machine. In one model, the program follows a qualitatively new kind of learning process in generating its own strategy and improving it on the basis of experience. The method enables the learner to exceed the playing quality of the teacher. It is suggested that the methods and techniques employed in the project may be useful in mechanizing some problem-solving activities that can be reduced to pattern recognition, such as meteorological forecasting, medical diagnosis, traffic control, and so on. No deliberate attempt has been made to imitate humans. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Additional Features of an Adaptive, Multicategory Pattern Classification System

    Publication Year: 1969 , Page(s): 183 - 191
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1839 KB)  

    Some additional features of an adaptive, multicategory pattern classification system are presented. No a priori knowledge of the class probability densities or a priori probabilities of occurrence of the categories is required. The system utilizes a set of functions selected by the user to form discriminant functions. Adaptation of the system is accomplished using a set of independent pattern samples of known classification in such a manner that the system discriminant functions form minimum mean-square approximations to the Bayes discriminant functions as the number of samples of known classification increases. The convergence rate of the system is examined, and conditions are established under which the expected loss due to misclassification by the system is asymptotically equivalent to the minimum loss achievable when using the Bayes discriminant functions. In addition, a simulation of the system for a three-category problem is presented to demonstrate system performance for a finite number of adaptions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Constraint Theory, Part III: Inequality and Discrete Relations

    Publication Year: 1969 , Page(s): 191 - 199
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1973 KB)  

    Parts I and II of this three-part paper provided the fundamental concepts underlying constraint theory whose goal is the systematic determination of whether a mathematical model and its computations are well posed. In addition to deriving results for the general relation, special relations defined as universal and regular were treated. This concluding part treats two more special relations: inequality and discrete. Employing the axiom of transitivity for inequalities, results relating to the consistency of a mathematical model of inequalities in terms of its model graph are derived. Rules for the simultaneous propagation of four types of constraint, over, point, interval, and slack, through a heterogeneous model graph are established. In contrast to other relation types, discrete relations point constrain every relevant variable, so that finding intrinsic constraint sources is trivial. A general procedure is provided to determine the allowability of requested computations on a discrete model. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Development of a Community Health Service System Simulation Model

    Publication Year: 1969 , Page(s): 199 - 207
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2216 KB)  

    The feasibility of using simulation techniques in community health planning by investigating problems associated with the development of a community health service system simulation model is examined. Two preliminary tasks were completed: a general systems analysis of the community health services and the preparation of a simulation model of one segment of this system. The system is described in terms of needs, demands, and resources. Various factors determine extent of the needs, conversion of the needs into demands, and availability of resources for satisfying the demands. This conceptual framework, the various factors, their significance, and an estimate of the extent of their control by a health planner were derived. A simulation model was developed using this framework and was based on a maternal and infant care program conducted by the North Carolina State Board of Health. The model was verified using data from three different communities within the program. Model processes are described; outputs are discussed; and potential model uses within a planning environment are presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stochastic Allocation and System Analysis

    Publication Year: 1969 , Page(s): 207 - 214
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1763 KB)  

    The optimum allocation of a fixed stock of unreliable units to a random number of demands is discussed. The demands occur at Poisson times; several types of criteria are described, but the most important optimum presented is of the probability that at least one allotted unit does not fail at every Poisson demand. Hence this concerns how unreliable elements can best be used to create a reliable system. The allocation problem arose in a military system analysis context. The results presented exemplify how system science concepts (Poisson models, recursive computation, and cost/ benefit comparisons) and current computing tools can be applied to practical problems. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Formulation of Fuzzy Automata and Its Application as a Model of Learning Systems

    Publication Year: 1969 , Page(s): 215 - 223
    Cited by:  Papers (33)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1592 KB)  

    Based on the concept of fuzzy sets defined by Zadeh, a class of fuzzy automata is formulated similar to Mealy's formulation of finite automata. A fuzzy automaton behaves in a deterministic fashion. However, it has many properties similar to that of stochastic automata. Its application as a model of learning systems is discussed. A nonsupervised learning scheme in automatic control and pattern recognition is proposed with computer simulation results presented. An advantage of employing fuzzy automaton as a learning model is its simplicity in design and computation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The Systems Approach to Test Evaluation

    Publication Year: 1969 , Page(s): 223 - 230
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2165 KB)  

    The evolution of large and sophisticated systems has created a requirement for extensive system tests programs. The argument that the change in scale of testing affects organizational communications so as to decrease the visibility provided by traditional management tools is presented. To attack this problem, a recommendation is made to establish a central test evaluation group with the primary goal of feeding back intelligence from the test program to the designers and managers. The concept of test "intelligence" is defined and a typical activity flow for the evaluation group is described. The general factors which determine the value of information are related to the specific variables of a test program. A rationale is developed for planning an evaluation approach based on the tradeoff between time invested in evaluation and the change in the value of the evaluation product (intelligence) as it ages. A functional approach to the reporting scheme is given. Report contents and timing are keyed to the program decision requirements. The potential benefits of a central, independent, and objective test evaluation are suggested for the designer, the manager, and the customer. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Electrochemically Active Field-Trainable Pattern Recognition Systems

    Publication Year: 1969 , Page(s): 230 - 237
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (5538 KB)  

    System performance tests have recently been carried out on an experimental electrochemically active/plastic system designed to demonstrate and begin to explore the theoretically predicted phenomenon we call temporally associative "field-trainability," a process which, when perfected and extended, may lead to revolutionary advances in manufacturing ultrahigh-density and highly versatile pattern recognition machines. The first complete experimental system, called-linear field-trainable, (LIFT) consists of two matching but opposing (excitatory/inhibitory) parallel arrays of active/plastic dipoles (iron/gold) in nitric acid, which are stimulated (electrically) in various patterns of activity. and respond by simple parallel coupling through the surrounding fluids and steel container walls to a low-impedance low-pass threshold detector. Alternatively, when desired, the chambers are connected to a power supply, the effect of which is to deliver massive electrical shocks to the entire array at once. The resulting reinforcement field is applied in synchronism with sample input pattern stimulations (which alternate with unreinforced response test stimulations) in attempts to induce at will, in a sequence of small steps, simultaneous changes of fine cellular structure which will produce corresponding specific systemic functional mutations as desired from among the class of linear decision functions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stochastic Learning of Time-Varying Parameters in Random Environment

    Publication Year: 1969 , Page(s): 237 - 246
    Cited by:  Papers (6)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2067 KB)  

    The problem of learning in nonstationary environment is formulated as that of estimating time-varying parameters of a probability distribution which characterizes the process under study. Dynamic stochastic approximation algorithms are proposed to estimate the unknown time-varying parameters in a recursive fashion. Both supervised and nonsupervised learning schemes are discussed and their convergence properties are investigated. An accelerated scheme for the possible improvement of the dynamic algorithm is given. Numerical examples and an application of the proposed algorithm to a problem in weather forecasting are presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recursive Estimates of Probability Densities

    Publication Year: 1969 , Page(s): 246 - 247
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (427 KB)  

    First Page of the Article
    View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Evolutionary Pattern Recognition Network

    Publication Year: 1969 , Page(s): 247 - 250
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (817 KB)  

    A pattern recognition network with two types of adaptation has been investigated. The network output is a weighted sum of the outputs of elements which compute real functions of the discrete network inputs. The first type of adaptation involves the adjustment of the weights while the second type involves the periodic replacement of the least valuable network elements with new ones. The expected error of the network in realizing arbitrary input-output functions has been found by Monte-Carlo simulation for simple weight adaptation and for the case where the population of network elements is allowed to evolve. Three heuristics for determining which elements are to be replaced in each generation have been evaluated and compared. These were based on the size of the weight associated with each element after training, a normalized weight size, and the cross correlation between the elemental function and the desired network function. All three selection criteria resulted in improvements of the network performance over the nonevolutionary case. The normalized weight size criterion was most effective while the cross-correlation criterion was least effective. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Every norm is not logarithmically convex

    Publication Year: 1969 , Page(s): 250 - 251
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (531 KB)  

    This correspondence relates to the remark in a recent paper by D.G. Luenberger [ibid., vol. SSC-4, pp. 182-188, July 1968] that any norm defined on a vector space is a real convex function. Although this is a well-known fact in mathematics, a less well-known fact is that every logarithmically convex function is positive and convex, but not conversely, i.e., there are positive convex functions which are not logarithmically convex. As the above title indicates, norms are such functions. This mathematical remark relates to systems science through several areas of application where logarithmic convexity is a highly useful property. In particular, Klinger and Mangasarian ["Logarithmic convexity and geometric programming," J. Math. Anal. and Appl., vol. 24, pp. 388-408, November 1968] mention optimization of multiplicative criteria, reliability theory, and electrical network synthesis, and examine geometric programming in detail. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of Optimal Control Theory to the Crashworthiness of a Passenger Vehicle Model

    Publication Year: 1969 , Page(s): 251 - 256
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1343 KB)  

    Optimal control theory concepts are thought to be useful in understanding the problem of determining safe deceleration characteristics for a crashing vehicle. These deceleration waveforms are to be computed such that passenger belt forces are minimized. Using both a linear one-degree-of-freedom model and a nonlinear two-degree-of-freedom model for a frontal collision, this problem is shown to be equivalent to the minimization of a performance or cost function when the terminal time is not fixed a priori, but is determined by terminal constraints. While the maximum principle is applied directly to find the optimal deceleration waveform for the linear problem, the steepest ascent method is used to optimize iteratively the nonlinear problem. Passenger seatbelt forces which resulted from using these optimal waveforms were compared with those forces which resulted from using step and ramp functions. Results showed that the seat belt forces resulting from the optimally derived deceleration signals were considerably smaller than those using step and ramp functions. With further effort, these results could possibly be used as design guides. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Strategy and a Tactic for Generation of Transportation Alternatives

    Publication Year: 1969 , Page(s): 257 - 260
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1146 KB)  

    Transportation planning strategy is discussed in its relationship to overall regional development policy in the Northeast Corridor. A tactic currently being applied in the Northeast Corridor Transportation Project to generate alternative transportation system descriptions is presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Correction to "On the Inverse of Linear Dynamical Systems"

    Publication Year: 1969 , Page(s): 260
    Save to Project icon | Request Permissions | PDF file iconPDF (231 KB)  
    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Book Reviews

    Publication Year: 1969 , Page(s): 260 - 262
    Save to Project icon | Request Permissions | PDF file iconPDF (788 KB)  
    Freely Available from IEEE
  • Contributors

    Publication Year: 1969 , Page(s): 263 - 264
    Save to Project icon | Request Permissions | PDF file iconPDF (1997 KB)  
    Freely Available from IEEE
  • Information for authors

    Publication Year: 1969 , Page(s): 264b
    Save to Project icon | Request Permissions | PDF file iconPDF (171 KB)  
    Freely Available from IEEE