New papers accepted

Two new papers from the group have been accepted for their publication:

The paper “On the Waterfall Performance of Finite-length SC-LDPC Codes Constructed from Protographs” by Markus Stinner and Pablo M. Olmos has been accepted for publication in IEEE Journal on Selected Areas in Communications, special issue on Recent Advances in Capacity Approaching codes. November 2015.
An analysis of spatially-coupled low-density parity-check (SC-LDPC) codes constructed from protographs is proposed. Given the protograph used to generate the SC-LDPC code ensemble, a set of scaling parameters to characterize the average finite-length performance in the waterfall region is computed. The error performance of structured SC-LDPC code ensembles is shown to follow a scaling law similar to that of unstructured randomly-constructed SC-LDPC codes. Under a finite-length perspective, some of the most relevant SC-LDPC protograph structures proposed to date are compared. The analysis reveals significant differences in their finite-length scaling behavior, which is corroborated by simulation. Spatially-coupled repeat-accumulate codes present excellent finite-length performance, as they outperform in the waterfall region SC-LDPC codes of the same rate and better asymptotic thresholds.
The paper “Infinite Continuous Feature Model for Psychiatric Comorbidity Analysis” by Isabel Valera, Francisco Ruiz, Pablo M. Olmos, Carlos Blanco and Fernando Pérez-Cruz has been accepted for publication in Neural Computation. October 2015.
We aim at finding the comorbidity patterns of substance abuse, mood and personality disorders using the diagnoses from the National Epidemiologic Survey on Alcohol and Related Conditions database. To this end, we propose a novel Bayesian nonparametric latent feature model for categorical observations, based on the Indian buffet process, in which the latent variables can take values between 0 and 1. The proposed model has several interesting features for modeling psychiatric disorders. First, the latent features might be off, which allows distinguishing between the subjects that suffer a condition and those who do not. Second, the active latent features take positive values, which allows modeling the extent to which the patient suffers that condition. We also develop a new MCMC inference algorithm for our model that makes use of a nested expectation propagation procedure.