Programa de Doctorat en Tecnologies de la Informació i les Comunicacions: Envíos recientes
Mostrando ítems 41-60 de 394
Assessing the digital well-being of educational technologies supported by learning analytics
The accelerating adoption of digital technologies creates a direct relationship between the status of individual and societal well-being on one hand, and the state of the digital technologies that underpin human life ...
Automatic characterization and generation of music loops and instrument samples for electronic music production
Repurposing audio material to create new music - also known as sampling - was a foundation of electronic music and is a fundamental component of this practice. Currently, large-scale databases of audio offer vast collections ...
Assessing the impact of music recommendation diversity on listeners
This thesis focuses on assessing the impact that music recommendation diversity may have on listeners. In the music domain, diversity is one of the values that recommender systems should preserve, because the world music ...
Risk assessment in complex data settings: algorithmic fairness and causal inference
We try to address some challenges in structured risk assessment tools in two application areas of recidivism risk in criminal justice and dropout risk in higher education domain. Using Machine Learning (ML) methods, we ...
Brain states in health and disease: insights from neuroimaging and theoretical neuroscience
Spontaneous brain activity persists and transitions between brain states, such as from wakefulness to sleep, from development to ageing, or it may transition to pathological states such as coma. Nevertheless, a consensual ...
Characterizing brain mechanisms underlying Neurofeedback and Meditation through whole-brain computational modeling and artificial intelligence
En las últimas décadas, un cambio de paradigma en el estudio y el tratamiento de los trastornos cerebrales ha hecho que se preste más atención a nuevas intervenciones terapéuticas. En esta línea, las técnicas de autorregulación, ...
Grounding brain mechanisms in the spatiotemporal properties of cerebral and behavioral networks
The spatiotemporal properties of the cerebral and behavioral networks are key elements to understand normal brain functioning and their breakdowns. This dissertation focuses on the cerebral networks of the brain regions’ ...
Algorithmic bias in graph-based recommender systems
Recommender Systems represent a key instrument to convey consumption of contents available on the Web. They enhance the engagement among the users and the online platforms through algorithmic personalization. Injecting ...
The Next Wi-Fi Generation: towards intelligent and multi-link enabled networks
The next Wi-Fi generation poses in front of a massive challenge as the main enabler for new services and applications. Traffic requirements are expected to keep rising year over year, challenging Wi-Fi networks to cope ...
Human-centered machine learning for music emotion recognition
This doctoral thesis is focused on music in terms of emotion -- such algorithms compose the computational task of music emotion recognition (MER). MER evaluates emotionally relevant features from music, correlates them ...
Coupling by volume conduction as a wireless power transfer and communications method for threadlike biomedical implants.
Although the use of volumetric conduction to power wireless implants has been experimentally validated in recent years, a theoretical framework to determine the power and efficiency obtained by minimally invasive implants ...
Algebraic techniques for universal succinct arguments
In this thesis, we make theoretical and practical contributions to the design of succinct arguments with universal setups in the pairing-based setting. We first introduce a new primitive, Checkable Subspace Sampling (CSS) ...
Symmetries in constraint satisfaction: Weisfeiler-Leman invariance and promise problems
This thesis focuses on the complexity of the fixed-template Constraint Satisfaction Problem (CSP) and its variants. Our contributions are two-fold. On the one hand, we study how closure of the space of CSP instances under ...
Computational fluid simulations in clinical datasets for understanding thrombus formation before and after left atrial appendage occlusion
Atrial fibrillation is considered the most common arrhythmia in humans. Because the left atrium of the heart stops beating properly and begins to do so in an arrhythmic manner, the blood may become stagnant in a small ...
Succinct arguments: efficiency, assumptions and trade-offs
Succinct non-interactive arguments (snarks) are cryptographic constructions that allow a prover to convince a verifier about the validity of a statement regarding some computation. We consider these objects from the ...
Towards audio-visual deep learning methods for singing voice separation and lip synchronization
A typical music audio consists of different sounds spread across voices and musical instruments. We propose a low-cost deep learning (DL) method for simultaneously isolating multiple sound sources present in a music ...
Decision-making during action
Decision making requires coordinating motor actions that are necessary to report the choices and sample relevant information from the environment. For instance, when riding a bike on a busy road, rider must both pedal while ...
Computational image analysis methods for the study of perinatal brain development
Perinatal medicine has drawn increasing attention by the neuroscientific research community, as an early detection of most pathologies leads to more effective therapies and treatments. Monitoring fetal and neonatal brain ...
Algorithmic fairness in practice: data, models and interactions
The fast-growing adoption of technologies based on Machine Learning (ML), in addition to the large scale at which they operate, makes them a potential source of systematic discrimination against disadvantaged social ...
Modeling timbre for neural singing synthesis: methods for data-efficient, reduced effort voice creation, and fast and stable inference
Singing synthesis has seen a notable surge in popularity in the last decade and a half. Music producers use this technology as an instrument, there is an audience for music with synthetic vocals, and an entire range of ...
