Properties of cross-entropy minimization

Comparison 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...

Optimization of Double Pipe Fin-pin Heat Exchanger Using Entropy Generation Minimization

In the current work optimization of double pipe fin-pin heat exchanger has been studied. The effective parameters those are controlling the heat exchanger performance are divided in two categories; geometrical and operating conditions. Using the Brent’s optimization algorithm and handling the thermodynamically design concept, one can minimize entropy generation in different length of heat excha...

Nonextensive triangle equality and other properties of Tsallis relative-entropy minimization

Kullback–Leibler relative-entropy has unique properties in cases involving distributions resulting from relative-entropy minimization. Tsallis relative-entropy is a one-parameter generalization of Kullback–Leibler relative-entropy in the nonextensive thermostatistics. In this paper, we present the properties of Tsallis relative-entropy minimization and present some differences with the classica...

A Logic for Default Reasoning About Probabilities

A logic is defined that allows to express in­ formation about statistical probabilities and about degrees of belief in specific proposi­ tions. By interpreting the two types of proba­ bilities in one common probability space, the semantics given are well suited to model the influence of statistical information on the for­ mation of subjective beliefs. Cross entropy minimization is a key element...

In supervised training, we minimize empirical loss over a training set:! prediction: ground-truth: loss function: Empirical risk minimization: Example for a classification problem: cross-entropy loss