A5070108326- Health, Medicine and Society
- Healthcare Systems and Practices
- Social Sciences and Governance
- Social Policies and Family
- European and International Contract Law
- Aging, Elder Care, and Social Issues
- Legal and Social Philosophy
- French Urban and Social Studies
- Historical and Scientific Studies
- Criminal Law and Evidence
- Ethics and Social Impacts of AI
- Adversarial Robustness in Machine Learning
- Explainable Artificial Intelligence (XAI)
- Corporate Governance and Law
- Migration, Identity, and Health
- Corporate Insolvency and Governance
- Comparative and International Law Studies
- Insurance and Financial Risk Management
- Agriculture and Rural Development Research
- Privacy-Preserving Technologies in Data
- Advanced Neural Network Applications
- Migration, Aging, and Tourism Studies
- Occupational Health and Safety Research
- Mental Health via Writing
- Parallel Computing and Optimization Techniques
Telefonica Research and Development
2023-2024
Institut Droit et Santé
2010-2024
Universitat Pompeu Fabra
2020-2023
Université Paris Cité
2020-2021
Universidad de Sevilla
2014
Social media sites are becoming an increasingly important source of information about mental health disorders. Among them, eating disorders complex psychological problems that involve unhealthy habits. In particular, there is evidence showing signs and symptoms anorexia nervosa can be traced in social platforms. Knowing input data biases tend to amplified by artificial intelligence algorithms and, machine learning, these methods should revised mitigate biased discrimination such domains.The...
Novel Deep Learning (DL) algorithms show ever-increasing accuracy and precision in multiple application domains. However, some steps further are needed towards the ubiquitous adoption of this kind instrument. First, effort skills required to develop new DL models, or adapt existing ones use-cases, hardly available for small- medium-sized businesses. Second, inference must be brought at edge, overcome limitations posed by classically-used cloud computing paradigm. This requires implementation...
Machine Learning (ML) techniques have been increasingly adopted by the real estate market in last few years. Applications include, among many others, predicting value of a property or an area, advanced systems for managing marketing and ads campaigns, recommendation based on user preferences. While these can provide important benefits to business owners users platforms, algorithmic biases result inequalities loss opportunities groups people who are already disadvantaged their access housing....
The accuracy and fairness of perception systems in autonomous driving are crucial, particularly for vulnerable road users. Mainstream research has looked into improving the performance metrics classification accuracy. However, hidden traits bias inheritance AI models, class imbalances disparities datasets often overlooked. In this context, our study examines users by focusing on distribution analysis, evaluation, impact assessment. We identify concern representation, leading to potential...
As Automatic Speech Recognition (ASR) models become ever more pervasive, it is important to ensure that they make reliable predictions under corruptions present in the physical and digital world. We propose Robust Bench (SRB), a comprehensive benchmark for evaluating robustness of ASR diverse corruptions. SRB composed 69 input perturbations which are intended simulate various may encounter use evaluate several state-of-the-art observe model size certain modeling choices such as discrete...
Training and deploying Machine Learning models that simultaneously adhere to principles of fairness privacy while ensuring good utility poses a significant challenge. The interplay between these three factors trustworthiness is frequently underestimated remains insufficiently explored. Consequently, many efforts focus on only two factors, neglecting one in the process. decentralization datasets variations distributions among clients exacerbate complexity achieving this ethical trade-off...
Artificial Intelligence (AI) is increasingly used to build Decision Support Systems (DSS) across many domains. This paper describes a series of experiments designed observe human response different characteristics DSS such as accuracy and bias, particularly the extent which participants rely on DSS, performance they achieve. In our experiments, play simple online game inspired by so-called "wildcat" (i.e., exploratory) drilling for oil. The landscape has two layers: visible layer describing...
The use of Deep Learning (DL) algorithms is increasingly evolving in many application domains. Despite the rapid growing algorithm size and complexity, performing DL inference at edge becoming a clear trend to cope with low latency, privacy bandwidth constraints. Nevertheless, traditional implementation on low-energy computing nodes often requires experience-based manual intervention trial-and-error iterations get functional effective solution. This work presents computer-aided design (CAD)...
Research in adversarial machine learning has shown how the performance of models can be seriously compromised by injecting even a small fraction poisoning points into training data. While effects on model accuracy such attacks have been widely studied, their potential other metrics remain to evaluated. In this work, we introduce an optimization framework for against algorithmic fairness, and develop gradient-based attack aimed at introducing classification disparities among different groups...
Artificial Intelligence (AI) and its relation with societies has become an increasingly interesting subject of study for the social sciences. Nevertheless, there is still important lack interdisciplinary empirical research applying theories to field AI. We here aim shed light on interactions between humans autonomous systems analyse moral conventions, which underly these cause moments conflict cooperation. For this purpose we employ Economics Convention (EC), originally developed in context...