Abstract.
Classical Csiszár-Kullback inequalities bound the L 1-distance of two probability densities in terms of their relative (convex) entropies. Here we generalize such inequalities to not necessarily normalized and possibly non-positive L 1 functions. Also, we analyse the optimality of the derived Csiszár-Kullback type inequalities and show that they are in many important cases significantly sharper than the classical ones (in terms of the functional dependence of the L 1 bound on the relative entropy). Moreover our construction of these bounds is rather elementary.
Similar content being viewed by others
Author information
Authors and Affiliations
Additional information
(Received 18 February 2000; in revised form 13 June 2000)
Rights and permissions
About this article
Cite this article
Unterreiter, A., Arnold, A., Markowich, P. et al. On Generalized Csiszár-Kullback Inequalitieys. Mh Math 131, 235–253 (2000). https://doi.org/10.1007/s006050070013
Issue Date:
DOI: https://doi.org/10.1007/s006050070013