María Pérez-Ortiz

Computer Scientist

I am a researcher in computer science with a passion for machine learning, AI and science dissemination. I am currently a postdoctoral Research Fellow at University College London (UK).

Contact me:
contact@mariaperezortiz.com

Who am I?

I'm a statistically rare mix of science and literature trying to embrace my juxtapositions. I was brought up between logic puzzles, chess and poetry. Now I am obsessed with AI, neuroscience and writing science fiction.

My research interests

Do more with less (data efficiency)

Use of synthetic data to complement real one, active sampling, semi-supervised learning...

Multimodal learning

Models that can process and relate data from different modalities (e.g. imaging and text).

Inclusion of human expertise on models

Priors, priors and priors. And imitation learning.

Weakly supervised learners

More data but less labels, or labels with high uncertainty.

Adaptive machine learning

Systems that adapt to the environment, reinforcement learning.

Automatisation of data science

How to create memory from datasets to guide the automated statistician.

Impact of AI on our society

Automatisation, changing job market, ethics...

Work Experience

2018 - Present

Research Fellow

Department of Computer Science, University College London (UK)

2017 - 2018

Research Associate

Department of Computer Science and Technology, University of Cambridge (UK)

2013/2017

Visiting academic

School of Computer Science, University of Birmingham (UK)

2015 - 2017

Lecturer

Department of Quantitative methods, University Loyola Andalucia (Spain)

2014 - 2015

Research assistant

Spanish National Research Council (Spain)

2011 - 2014

Research assistant

Department of Computer Science and Numerical Analysis, University of Cordoba (Spain)

Education

2012 - 2015

PhD in Machine Learning

University of Cordoba (Spain)

2011 - 2012

MSc (Hons) in Intelligent Systems

University of Cordoba (Spain)

2008 - 2011

BSc in Computer Science

University of Cordoba (Spain)

2015 - 2016

Certified Health Coach

Institute of Integrative Nutrition (US)

Other interests

Writing

I love writing and I try to contribute to science dissemination in press.

Dancing

I have completed the first part of the dance conservatoire, focusing on Ballet and flamenco. Now I am interested in the use of dancing as a mean of meditation and mindfulness.

Health coaching

I also hold a master on health coaching from the Institute for Integrative Nutrition in the US.

Teaching innovation

I am very interested in the use of humour, mindfulness, coaching and gaming in teaching environments. Also, on how to bring transversal topics to the classroom, such as nutrition.

Selected publications

  • Oversampling the Minority Class in the Feature Space. M. Perez-Ortiz, PA Gutierrez, P. Tino and C. Hervas-Martinez, IEEE Transactions on Neural Networks and Learning Systems, 8, 2015.

  • Graph-based approaches for over-sampling in the context of ordinal regression. M. Perez-Ortiz, PA Gutierrez, C. Hervas-Martinez, X. Yao, IEEE Transactions on Knowledge and Data Engineering 27 (5), 1233-1245, 2015.

  • Projection-Based Ensemble Learning for Ordinal Regression, M. Perez-Ortiz, PA Gutierrez and C. Hervas-Martinez, IEEE Transactions on Cybernetics, 2013.

  • Ordinal regression methods: survey and experimental study. PA Gutiérrez, M. Perez-Ortiz, J. Sanchez-Monedero, F. Fernandez-Navarro and C. Hervas-Martinez, IEEE Transactions on Knowledge and Data Engineering, 28(1), 127-146, 2016.

  • Semi-supervised learning for ordinal Kernel Discriminant Analysis. M. Perez-Ortiz, PA Gutierrez, M. Carbonero-Ruz and C. Hervas-Martinez, Neural Networks 84, 57-66, 2016.

  • A study on multi-scale kernel optimisation via centered kernel-target alignment. M. Perez-Ortiz, PA Gutierrez, J. Sanchez-Monedero and C. Hervas-Martinez, Neural Processing Letters 44 (2), 491-517, 2016.

  • An organ allocation system for liver transplantation based on ordinal regression. M. Perez-Ortiz, M. Cruz-Ramirez, M.D. Ayllon-Teran, N. Heaton, R. Ciria and C. Hervas-Martinez, Applied Soft Computing 14, 88-98, 2014.

  • A semi-supervised system for weed mapping in sunflower crops using unmanned aerial vehicles and a crop row detection method. M. Perez-Ortiz, JM Pena, PA Gutierrez, J. Torres-Sanchez, F. Lopez-Granados and C. Hervas-Martinez, Applied Soft Computing, 37, 533-544, 2015.

  • Synthetic semi-supervised learning in imbalanced domains: Constructing a model for donor-recipient matching in liver transplantation. M. Perez-Ortiz, PA Gutierrez, MD Ayllon-Teran, N. Heaton, R. Ciria and C. Hervas-Martinez, Knowledge-Based Systems 123, 75-87, 2017.

  • On the use of evolutionary time series analysis for segmenting paleoclimate data. M. Perez-Ortiz, A.M. Duran-Rosal, PA Gutierrez, J. Sanchez-Monedero, A. Nikolaou, F. Fernandez-Navarro and C. Hervas-Martinez, Neurocomputing, 2017.

  • Incorporating privileged information to improve manifold ordinal regression. M. Perez-Ortiz, PA Gutierrez and C. Hervas-Martinez, International Conference on Neural Computation Theory and Applications (NCTA), 187-194, 2014.

  • Adapting Linear Discriminant Analysis to the Paradigm of Learning from Label Proportions. M. Perez-Ortiz, PA Gutierrez, M. Carbonero-Ruz and C. Hervas-Martinez, IEEE Symposium Series on Computational Intelligence (IEEE SSCI), 2016.

  • Binary Ranking for Ordinal Class Imbalance. R. Cruz, K. Fernandes, J.F. Pinto-Costa, M. Perez-Ortiz and J.S. Cardoso, Pattern Analysis and Applications, 2018.

  • Partial order label decomposition approaches for melanoma diagnosis. J. Sanchez-Monedero, M. Perez-Ortiz, A. Saez, P.A. Gutierrez and C. Hervas-Martinez, Applied Soft Computing 64, 341-355, 2018.

  • A practical guide and software for analysing pairwise comparison experiments. M. Perez-Ortiz and R. K. Mantiuk. In: arXiv pre-print, 2017, (project page).

  • A mixture of experts model for predicting persistent weather patterns. M. Perez-Ortiz, P. A. Gutierrez, P. Tino, C. Casanova-Mateo and S. Salcedo-Sanz, IEEE International Joint Conference on Neural Networks (IJCNN), 2018.

  • Psychometric scaling of TID2013 dataset. A. Mikhailiuk, M. Perez-Ortiz and R. K. Mantiuk, International Conference on Quality of Multimedia Experience (QoMex), 2018.

  • Trained perceptual transform for quality assessment of high dynamic range images and video. N. Ye, M. Perez-Ortiz and R. K. Mantiuk, International Conference on Image Processing (IEEE ICIP), 2018.

  • Exploiting synthetically generated data with semi-supervised learning for small and imbalanced datasets. M. Perez-Ortiz, P. Tino, R. K. Mantiuk and C. Hervas-Martinez, AAAI Conference on Artificial Intelligence, 2019.

Contact

Address
UCL Computer Science, 66-72 Gower Street
London (UK)
Email
contact@mariaperezortiz.com