1 code implementation • 8 Jun 2023 • Gonzalo Martínez, Lauren Watson, Pedro Reviriego, José Alberto Hernández, Marc Juarez, Rik Sarkar
Our results show that the quality and diversity of the generated images can degrade over time suggesting that incorporating AI-created data can have undesired effects on future versions of generative models.
no code implementations • 17 Feb 2023 • Gonzalo Martínez, Lauren Watson, Pedro Reviriego, José Alberto Hernández, Marc Juarez, Rik Sarkar
Therefore, future versions of generative AI tools will be trained with Internet data that is a mix of original and AI-generated data.
no code implementations • 8 Sep 2022 • Marc Juarez, Samuel Yeom, Matt Fredrikson
Our experimental results on real-world datasets show that this approach is effective, achieving 80--100% AUC-ROC in detecting shifts involving the underrepresentation of a demographic group in the training set.
1 code implementation • 24 Jun 2022 • Marc Juarez, Aleksandra Korolova
As in traditional machine learning models, models trained with federated learning may exhibit disparate performance across demographic groups.
no code implementations • 24 Jun 2019 • Sandra Siby, Marc Juarez, Claudia Diaz, Narseo Vallina-Rodriguez, Carmela Troncoso
Virtually every connection to an Internet service is preceded by a DNS lookup which is performed without any traffic-level protection, thus enabling manipulation, redirection, surveillance, and censorship.
Cryptography and Security
no code implementations • 21 Aug 2017 • Vera Rimmer, Davy Preuveneers, Marc Juarez, Tom Van Goethem, Wouter Joosen
In our open world evaluation, the most performant deep learning model is 2% more accurate than the state-of-the-art attack.