Error-Entropy Minimization for Dynamical Systems Modeling
نویسنده
چکیده
Recent publications have presented many successful usages of elements from information theory in adaptive systems training. Errorentropy has been proven to outperform mean squared error as a cost function in many artificially generated data sets, but still few applications to real world data have been published. In this paper, we design a neural network trained with error-entropy minimization criterion and use it for dynamical systems modeling on artificial as well as real world data. Performance of this neural network is compared against the mean squared error driven approach in terms of computational complexity, parameter optimization, and error probability densities.
منابع مشابه
Entropy operator for continuous dynamical systems of finite topological entropy
In this paper we introduce the concept of entropy operator for continuous systems of finite topological entropy. It is shown that it generates the Kolmogorov entropy as a special case. If $phi$ is invertible then the entropy operator is bounded with the topological entropy of $phi$ as its norm.
متن کاملObservational Modeling of the Kolmogorov-Sinai Entropy
In this paper, Kolmogorov-Sinai entropy is studied using mathematical modeling of an observer $ Theta $. The relative entropy of a sub-$ sigma_Theta $-algebra having finite atoms is defined and then the ergodic properties of relative semi-dynamical systems are investigated. Also, a relative version of Kolmogorov-Sinai theorem is given. Finally, it is proved that the relative entropy of a...
متن کاملENTROPY OF DYNAMICAL SYSTEMS ON WEIGHTS OF A GRAPH
Let $G$ be a finite simple graph whose vertices and edges are weighted by two functions. In this paper we shall define and calculate entropy of a dynamical system on weights of the graph $G$, by using the weights of vertices and edges of $G$. We examine the conditions under which entropy of the dynamical system is zero, possitive or $+infty$. At the end it is shown that, for $rin [0,+infty]$, t...
متن کاملComparison of entropy generation minimization principle and entransy theory in optimal design of thermal systems
In this study, the relationship among the concepts of entropy generation rate, entransy theory, and generalized thermal resistance to the optimal design of thermal systems is discussed. The equations of entropy and entransy rates are compared and their implications for optimization of conductive heat transfer are analyzed. The theoretical analyses show that based on entropy generation minimizat...
متن کاملThe concept of logic entropy on D-posets
In this paper, a new invariant called {it logic entropy} for dynamical systems on a D-poset is introduced. Also, the {it conditional logical entropy} is defined and then some of its properties are studied. The invariance of the {it logic entropy} of a system under isomorphism is proved. At the end, the notion of an $ m $-generator of a dynamical system is introduced and a version of the Kolm...
متن کامل