Maximizing Rényi entropy rate
Christoph Bunte, Amos Lapidoth
Abstract
Christoph Bunte, Amos Lapidoth
Abstract
Of all univariate distributions on the nonnegative reals of a given mean, the distribution that maximizes the Rényi entropy is Lomax. But the memoryless Lomax stochastic process does not maximize the Rényi entropy rate: For Rényi orders smaller than one the supremum of the Rényi entropy rates is infinite, and for orders larger than one it is the differential Shannon entropy of the exponential distribution, which is the distribution that maximizes the differential Shannon entropy subject to these constraints. This is shown to be a special case of a much more general principle.
OpenAlex reports 3 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Of all univariate distributions on the nonnegative reals of a given mean, the distribution that maximizes the Rényi entropy is Lomax. But the memoryless Lomax stochastic process does not maximize the Rényi entropy rate: For Rényi orders smaller than one the supremum of the Rényi entropy rates is infinite, and for orders larger than one it is the differential Shannon entropy of the exponential distribution, which is the distribution that maximizes the differential Shannon entropy subject to these constraints. This is shown to be a special case of a much more general principle.
Key concepts: Mathematics, Rényi entropy, Min entropy, Entropy rate, Differential entropy, Maximum entropy probability distribution, Entropy (arrow of time), Applied mathematics