M
Marius Paraschiv
Researcher at IMDEA
Publications - 7
Citations - 41
Marius Paraschiv is an academic researcher from IMDEA. The author has contributed to research in topics: Shapley value & Computer science. The author has an hindex of 2, co-authored 5 publications receiving 15 citations.
Papers
More filters
Proceedings ArticleDOI
Mis-shapes, Mistakes, Misfits: An Analysis of Domain Classification Services
Pelayo Vallina,Victor Le Pochat,Álvaro Feal,Marius Paraschiv,Julien Gamba,Tim Burke,Oliver Hohlfeld,Juan E. Tapiador,Narseo Vallina-Rodriguez +8 more
TL;DR: This study empirically explores popular domain classification services, their methodologies, scalability limitations, label constellations, and their suitability to academic research as well as other practical applications such as content filtering, and concludes with actionable recommendations on their usage.
Journal ArticleDOI
Classification of Underwater Fish Images and Videos via Very Small Convolutional Neural Networks
Marius Paraschiv,Ricardo Padrino,Paolo Casari,Eyal Bigal,Aviad Scheinin,Dan Tchernov,Antonio Fernández Anta +6 more
TL;DR: This paper proposes small convolutional neural networks as a practical engineering solution that helps tackle fish image classification, and shows that even networks with little more than 12,000 parameters provide an acceptable working degree of accuracy in the classification task, even when trained on small and unbalanced datasets.
Posted Content
Valuating User Data in a Human-Centric Data Economy.
TL;DR: This paper demonstrates how a Human-Centric Data Economy would compensate the users of an online streaming service by borrowing the notion of the Shapley value from cooperative game theory to define what a fair compensation for each user should be for movie scores offered to the recommender system of the service.
Proceedings ArticleDOI
Very Small Neural Networks for Optical Classification of Fish Images and Videos
TL;DR: In this paper, a small convolutional network is used to process images as well as videos frames for the classification of multiple species of pelagic fish. But the network requires a considerable amount of training data to produce meaningful results, which restricts their deployment on low power embedded field equipment.
Posted Content
Computing the Value of Spatio-Temporal Data in Wholesale and Retail Data Marketplaces.
TL;DR: It is shown that the relative value of the data held by different taxi companies and drivers may differ substantially, and that its relative ranking may change from district to district within a metropolitan area.