Title: “Nested filtering methods for Bayesian inference in state space models” Advisor: Joaquín Míguez Arenas. ABSTRACT A common feature to many problems in some of the most active fields of science is the need to calibrate (i.e., estimate the parameters) and then forecast the time evolution of high-dimensional dynamical systems using sequentially collected data. In…

Title Medical Data Wrangling With Sequential Variational Autoencoders Authors D. Barrejón, Pablo M. Olmos and Antonio Artés-Rodríguez Abstract Medical data sets are usually corrupted by noise and missing data. These missing patterns are commonly assumed to be completely random, but in medical scenarios, the reality is that these patterns occur in bursts due to sensors…

In this aid, the beneficiary is Alejandro Cano Sánchez and the abstract of the project is the following. The project is focused on the analysis and development of advanced methods for orbital uncertainty characterization applied to space surveillance and tracking of space debris, including the following aspects: modelling of the uncertainty in dynamical models and…

In this aid, the beneficiary is Arturo Armario Romero and the abstract of the project is the following. The goal of this project, named as ESPECTRO, is to provide eB2 with cutting-edge tools to combine and learn from multi-modal information. On the one hand, by providing classical methods with larger expressivity using BNP/ARD priors combined…

Authors Pablo Moreno-Muñoz, Antonio Artes-Rodríguez, Mauricio Álvarez Abstract We present a framework for transfer learning based on modular variational Gaussian processes (GP). We develop a module-based method that having a dictionary of well fitted GPs, each model being characterised by its hyperparameters, pseudo-inputs and their corresponding posterior densities, one could build ensemble GP models without…

Area of research: computational statistics Keywords: Bayesian computation, Monte Carlo, particle filtering, high-dimensional dynamical models, machine learning, aerospace engineering, e-health. Duration: 3+1 years Gross salary: 17,000€ per year Contact: Prof. Joaquin Miguez (joaquin.miguez@uc3m.es) We are opening two positions for PhD students within the Signal Processing & Learning Group. Candidates should have a background in mathematics,…

Area of research: computational statistics Keywords: Bayesian computation, Monte Carlo, particle filtering, high-dimensional dynamical models, machine learning, aerospace engineering, e-health. Duration: 1+1 years Gross salary: 24,000€ per year Contact: Prof. Joaquin Miguez (joaquin.miguez@uc3m.es) We are opening a postdoctoral position within the Signal Processing & Learning Group, to conduct research in problems related to Bayesian estimation,…

September 2020 The article “Actigraphic recording of motor activity in depressed inpatients: A novel computational approach to prediction of clinical course and hospital discharge” by Ignacio Peis, Javier-David Lopez-Morinigo, M. Mercedes Perez-Rodriguez, Maria-Luisa Barrigon, Marta Ruiz-Gomez, Antonio Artés-Rodríguez and Enrique Baca-García has been accepted for publication in Scientific Reports. Title: Actigraphic recording of motor activity in depressed inpatients: A…

August 2020 The researchers Antonio Artés Rodríguez and Gonzalo Ríos Muñoz at the Universidad Carlos III de Madrid (UC3M) together with the Fundación para la Investigación Biomédica del Hospital Gregorio Marañón (FIBHGM), have patented a new system “Sistema y método para la detección automática de patrones electrofisiológicos anómalos”, that allows a real time automatic detection…

July 2020 Congratulations to Daniel Barrejón, winner of a Predoctoral Fellowship of the highly competitive program FPU (Formación Personal Universitario) from the Spanish Government (Ministry of Education). He will be investigating in the Electronic and Communications Engineering field during the next 4 years at the Universidad Carlos III de Madrid (UC3M).