RESUMO
Proximal Markov Chain Monte Carlo is a novel construct that lies at the intersection of Bayesian computation and convex optimization, which helped popularize the use of nondifferentiable priors in Bayesian statistics. Existing formulations of proximal MCMC, however, require hyperparameters and regularization parameters to be prespecified. In this work, we extend the paradigm of proximal MCMC through introducing a novel new class of nondifferentiable priors called epigraph priors. As a proof of concept, we place trend filtering, which was originally a nonparametric regression problem, in a parametric setting to provide a posterior median fit along with credible intervals as measures of uncertainty. The key idea is to replace the nonsmooth term in the posterior density with its Moreau-Yosida envelope, which enables the application of the gradient-based MCMC sampler Hamiltonian Monte Carlo. The proposed method identifies the appropriate amount of smoothing in a data-driven way, thereby automating regularization parameter selection. Compared with conventional proximal MCMC methods, our method is mostly tuning free, achieving simultaneous calibration of the mean, scale and regularization parameters in a fully Bayesian framework.
RESUMO
Por la imposibilidad de usar convenientemente el tesauro Descriptores en Ciencias de la Salud (DeCS) en la asignación de epígrafes durante el proceso de catalogación, se elabora un conjunto de instrucciones técnicas, encaminadas a posibilitar que los bibliotecarios del Sistema Nacional de Información de Ciencias Médicas puedan utilizarlo como epigrafiario, dado que, en su mayoría, éstos no disponen de sistemas automatizados para el procesamiento de la información, por lo que sustentan el procedimiento de búsqueda y recuperación en la ficha catalográfica. Se aborda la representación temática de los documentos mediante epígrafes y sus principios básicos, así como las reglas para la coordinación de epígrafes y subepígrafes y las restricciones para el empleo de algunos tipos de encabezamientos. Se recomienda la aplicación del anexo Práctica catalográfica en el DeCS, preparado sobre la base del Cataloging practices del MeSH anotado, como estrategia para eliminar las contradicciones generadas por la ausencia de disposiciones adecuadas para representar todos los aspectos abordados en los documentos, sujetos a la asignación de epígrafes y subepígrafes.
Due to the failure of using adequately the thesaurus «Descriptores en Ciencias de la Salud (DeCS)» in the assignment of epigraphs during the cataloging process, a number of technical instructions are drawn. These instructions enable the librarians from the National System of Medical Sciences Information to use it as a dictionary of epigraphs because most of them do not have automatic systems for the processing of information so that they support themselves in the search and recovery procedures on the cataloging index card. The subject representation of documents through epigraphs and its basic principles are discussed as well as the rules for the coordination of epigraphs and subepigraphs and the restrictions in using some of the types of subheadings. The implementation of the annex Práctica catalográfica en el DeCS, based upon the Cataloging Practices of MeSH is recommended as a strategy to eliminate the contradictions generated by the lack of appropiate dispositions to represent all the aspects discussed in the documents and submitted to the assignment of epigraphs and subepigraphs.