A short bio

I am a lecturer and researcher at the Department of Mathematics and Computer Science of the University of Barcelona. I am interested in machine learning problems in which data uncertainty/missingness is determining characteristic, such as weakly supervised learning. I usually deal with this type of problems by means of probabilistic graphical models. Our proposals aim to learn probabilistic models, commonly Bayesian networks, from this kind of data.

I collaborate with Jesus Cerquides from the Artificial Intelligence Research Institute, where I spent most of my postdoctoral time. I completed my PhD studies with the Intelligent Systems Group of the University of the Basque Country under the supervision of Iñaki Inza and Jose A. Lozano.

Publications

  • Oct2019

    Variational importance sampling: initial findings

    J. Hernández-González, J. Capdevila, J. Cerquides

    In Proceedings of the 22nd International Conference of the Catalan Association for Artificial Intelligence (CCIA), 2019

  • Apr2019

    A framework for evaluation in learning from label proportions

    J. Hernández-González

    In Progress in Artificial Intelligence 8 (3): 359-373, 2019 · Additional data external link

  • Apr2019

    Beach litter forecasting on the SE coast of the Bay of Biscay: A bayesian networks approach

    I. Granado, O.C. Basurko, A. Rubio, L. Ferrer, J. Hernández-González, I. Epelde, J.A. Fernandes

    In Continental Shelf Research 180: 14-23, 2019

  • Jan2019

    Aggregated outputs by linear models: An application on marine litter beaching prediction

    J. Hernández-González, I. Inza, I. Granado, O.C. Basurko, J.A. Fernandes, J.A. Lozano

    In Information Sciences 481: 381-393, 2019 · Additional data external link

  • Oct2018

    Evaluation in learning from label proportions: an approximation to the precision-recall curve

    J. Hernández-González

    In Proceedings of the 18th Conference of the Spanish Association for Artificial Intelligence (CAEPIA), 2018 · Additional data external link

  • Oct2018

    Crowd learning with candidate labeling: an EM-based solution

    I. Beñaran-Muñoz, J. Hernández-González, A. Párez

    In Proceedings of the 18th Conference of the Spanish Association for Artificial Intelligence (CAEPIA), 2018 · Additional data external link

  • Jun2018

    A note on the behavior of majority voting in multi-class domains with biased annotators

    J. Hernández-González, I. Inza, J.A. Lozano

    In IEEE Transactions on Knowledge and Data Engineering 31(1): 195-200, 2018 · Additional data external link

  • Jan2018

    Two datasets of defect reports labeled by a crowd of annotators of unknown reliability

    J. Hernández-González, D. Rodriguez, I. Inza, R. Harrison, J.A. Lozano

    In Data in Brief 18: 840-845, 2018

  • Jan2018

    Learning to classify software defects from crowds: A novel approach

    J. Hernández-González, D. Rodriguez, I. Inza, R. Harrison, J.A. Lozano

    In Applied Soft Computing Journal 62: 579-591, 2018 · Additional data external link

  • Aug2017

    Merging knowledge bases in different languages

    J. Hernández-González, Estevam R. Hruschka Jr., Tom M. Mitchell

    In Proceedings of the 11th TextGraphs Workshop at ACL, 2017

  • Feb2017

    Learning from proportions of positive and unlabeled examples

    J. Hernández-González, I. Inza, J.A. Lozano

    International Journal of Intelligent Systems 32: 109--133, 2017

  • Sep2016

    Whatever you know, just tell me something: Crowd learning with free supervision

    J. Hernández-González, I. Inza, J.A. Lozano

    In Proceedings of the VIII Symposium of Data Mining Theory and Applications (TAMIDA), 2016

  • May2016

    Fitting the data from embryo implantation prediction: learning from label proportions

    J. Hernández-González, I. Inza, L. Crisol-Ortiz, M.A. Guembe, M.J. Iñarra, J.A. Lozano

    Statistical Methods in Medical Research 27(4): 1056-1066, 2018 · Additional data external link

  • Jan2016

    Weak supervision and other non-standard classification problems: a taxonomy

    J. Hernández-González, I. Inza, J.A. Lozano

    Pattern Recognition Letters 69: 49-55, 2016

  • Nov2015

    A novel weakly supervised problem: Learning from positive-unlabeled proportions

    J. Hernández-González, I. Inza, J.A. Lozano

    In Proceedings of the 16th Conference of the Spanish Association for Artificial Intelligence (CAEPIA), 2015

  • Jan2015

    Multidimensional learning from crowds: usefulness and application of expertise detection

    J. Hernández-González, I. Inza, J.A. Lozano

    International Journal of Intelligent Systems 30(3): 326-354, 2015

  • Dec2013

    Learning Bayesian network classifiers from label proportions

    J. Hernández-González, I. Inza, J.A. Lozano

    Pattern Recognition 46(12): 3425-3440, 2013 · Additional data external link

  • Sep2013

    Learning from crowds in multi-dimensional classification domains

    J. Hernández-González, I. Inza, J.A. Lozano

    In Proceedings of the 15th Conference of the Spanish Association for Artificial Intelligence (CAEPIA), 2013

  • Nov2011

    Learning naive Bayes models for Multiple-Instance Learning with label proportions

    J. Hernández-González, I. Inza

    In Proceedings of the 14th Conference of the Spanish Association for Artificial Intelligence (CAEPIA), 2011

Special mentions

  • 2018

    Best PhD thesis award

    University of the Basque Country

  • 2016

    Best idea award

    Open Data Euskadi

  • 2015

    Best student paper award

    16th Conference of the Spanish Association for Artificial Intelligence (CAEPIA)