“…Such strategy increases the population's resistance to strong environmental changes. [14,15,17] revealed that in situations of good separation of the categories (i.e. small classification error), they involved long training intervals that caused excessive growth of weights magnitudes which in turn, diminish the re-adaptation ability to learn a new classification.…”
Section: Modeling Of Crime Rate Dynamicsmentioning
confidence: 99%
“…In our analysis of adaptation characteristics of the populations of intelligent agents, we utilize two different classification task models as introduced in [6,14,15]. If an agent is unable to solve a new recognition task within a limited time, it will "perish", be removed from the population and replaced by a newborn.…”
Section: The Single Layer Perceptron and Its Training Peculiaritiesmentioning
Abstract. We describe a new point of view that uses contextual information of the interrelated conditions in which criminality or deviant occurs by means of computer simulation to examine its role in a progressively changing society. Based on the functionalist point of view, we model society as a multi agent system and criminality as a noise. A noise is injected as training signals to the single layer perceptrons that functions as agents (i.e. society) and all agents must comply the fitness function. Failure to comply will result in the agent being removed from the "society" and be replaced by a "newborn" that inherits some "upbringing" information from its "mother's agent". Our simulation studies point toward the constructive effects of criminality. The major contribution in our study is that new paradigm can show and measure criminality effects as in contrast to mere verbal descriptions of the social sciences. The new paradigm could be used to provide the functional explanation of atypical trends in criminality in East-and Middle European and some other countries and provide "artificial experimental data" for future studies.
“…Such strategy increases the population's resistance to strong environmental changes. [14,15,17] revealed that in situations of good separation of the categories (i.e. small classification error), they involved long training intervals that caused excessive growth of weights magnitudes which in turn, diminish the re-adaptation ability to learn a new classification.…”
Section: Modeling Of Crime Rate Dynamicsmentioning
confidence: 99%
“…In our analysis of adaptation characteristics of the populations of intelligent agents, we utilize two different classification task models as introduced in [6,14,15]. If an agent is unable to solve a new recognition task within a limited time, it will "perish", be removed from the population and replaced by a newborn.…”
Section: The Single Layer Perceptron and Its Training Peculiaritiesmentioning
Abstract. We describe a new point of view that uses contextual information of the interrelated conditions in which criminality or deviant occurs by means of computer simulation to examine its role in a progressively changing society. Based on the functionalist point of view, we model society as a multi agent system and criminality as a noise. A noise is injected as training signals to the single layer perceptrons that functions as agents (i.e. society) and all agents must comply the fitness function. Failure to comply will result in the agent being removed from the "society" and be replaced by a "newborn" that inherits some "upbringing" information from its "mother's agent". Our simulation studies point toward the constructive effects of criminality. The major contribution in our study is that new paradigm can show and measure criminality effects as in contrast to mere verbal descriptions of the social sciences. The new paradigm could be used to provide the functional explanation of atypical trends in criminality in East-and Middle European and some other countries and provide "artificial experimental data" for future studies.
“…In [18] it was demonstrated that training speed of the single layer perceptron (SLP) based classifier depends on difference s = |t 1 -t 2 | between desired outputs, t 1 , t 2 , corresponding to two diverse pattern classes. The training speed was measured as a number of training epochs required to train SLP.…”
Abstract. The paper considers supervised learning algorithm of nonlinear perceptron with dynamic targets adjustment which assists in faster learning and cognition. A difference between targets of the perceptron corresponding to objects of the first and second categories is associated with stimulation strength. A feedback chain that controls the difference between targets is interpreted as synthetic emotions. In a population of artificial agents that ought to learn similar pattern classification tasks, presence of the emotions helps a larger fraction of the agents to survive. We found that optimal level of synthetic emotions depends on difficulty of the pattern recognition task and requirements to learning quality and confirm Yerkes-Dodson law found in psychology.
“…To ensure that the crisis managers of VRT are capable of doing just that, more complex crisis situations thus need to be trained more often, for example, by means of a digital serious game. The experienced stress level of the crisis managers should be kept at medium intensity during training, which is where fastest learning can be expected according to Raudys and Justickis (2003), along with best performance as implied by the Yerkes-Dodson law. Knowing the player's current level of experienced stress would thus allow manipulation of the serious game to ensure that players experience the optimal stress level for learning, for example, by adjusting the pace of the game.…”
Section: Analytical Skills and Stress At Veiligheidsregio Twentementioning
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.