Publications and talks

Research awards

Microsoft Research Dissertation Grant, 2018.

1st Place Innovation Award on Data Science at Data for Policy, 2016.

Best Student Presentation Award at Domestic Nuclear Detection Office (DNDO) Academic Research Initiative Grantees Conference, 2015.

Papers

Maria De-Arteaga, Alexey Romanov, Hanna Wallach, Jennifer Chayes, Christian Borgs, Alexandra Chouldechova, Sahin Geyik, Krishnaram Kenthapadi, Adam Kalai, “Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting”, ACM Conference on Fairness, Accountability, and Transparency (ACM FAT*), 2019. [To appear]

Maria De-Arteaga, Artur Dubrawski, Alexandra Chouldechova, “Learning under selective labels in the presence of expert consistency”, Workshop on Fairness, Accountability, and Transparency in Machine Learning (FAT/ML), 2018. [arXiv]

Maria De-Arteaga, Peter Huggins, Jonathan Elmer, Gilles Clermont, Artur Dubrawski, “Predicting Neurological Recovery with Canonical Autocorrelation Embeddings”, In submission.

Maria De-Arteaga, William Herlands, Daniel B. Neill, Artur Dubrawski, “Machine Learning for the Developing World”, ACM Transactions on Management Information Systems, 2018. [ACM]

Maria De-Arteaga, Artur Dubrawski, “Discovery of Complex Anomalous Patterns of Sexual Violence in El Salvador”, Data for Policy, 2016. [Zenodo]

Maria De­-Arteaga, Artur Dubrawski, Peter Huggins, “Canonical Autocorrelation Analysis for Radiation Threat Detection”.  Heinz First Paper, Heinz College / Data Analysis Project, Machine Learning Department, 2016. Carnegie Mellon University. [CMU]

William Herlands, Maria De-­Arteaga, Daniel Neill, and Artur Dubrawski, “Lass0: Sparse Non­Convex Regression by Local Search”, Neural Information Processing Systems (NIPS), Optimization for Machine Learning Workshop, 2015. [arXiv]

Maria De­-Arteaga, Ivan Eggel, Charles Kahn and Henning Müller, “Analyzing Image Search Behaviour of Radiologists: Semantics and Prediction of Query Results” Journal of Digital Imaging, 2015. [Springer]

Maria De­-Arteaga, Ivan Eggel, Bao H. Do, Daniel Rubin, Charles E. Kahn Jr. and Henning Müller, “Comparing Image Search Behaviour in the ARRS GoldMiner Search Engine and a Clinical PACS/RIS” Journal of Biomedical Informatics, 2015. [Elsevier]

Alejandro Riveros, Maria De-­Arteaga, Fabio A. Gonzalez, Sergio Jimenez and Henning Müller, “MindLab­UNAL: Comparing Metamap and T­-mapper for Medical Concept Extraction in SemEval 2014 Task 7”, SemEval, 2014. [ACL]

Maria De-­Arteaga, Sergio Jimenez, Julia Baquero, George Dueñas and Sergio Mancera, “Author Profiling Using Corpus Statistics, Lexicons and Stylistic Features” 10th PAN Evaluation Lab on Uncovering Plagiarism, Authorship. and Social Misuse, CLEF, 2013. [UniWeimar]

Maria De­-Arteaga, “Vector models in text mining and its applications to author profiling tasks” (in Spanish), Undergraduate thesis, 2013. [PDF]

Talks

“Challenges of data-driven decision making with humans in the loop”, Google Fairness in ML Workshop, 2018.

“Machine learning, sexual violence crimes and decision-support systems”, Office of the Attorney General of Colombia, 2018.

“Learning under selective labels in the presence of expert consistency”, Workshop on Fairness, Accountability, and Transparency in Machine Learning (FAT/ML), 2018.

“Machine Learning for the Developing World”, Data science and machine learning for development and humanitarian action session, UNESCO Tech4Dev Conference, 2018. Closing Keynote.

“Machine Learning for the Developing World”, École polytechnique fédérale de Lausanne (EPFL), 2018.

“Using expert consensus to learn under selective labels”, Google Women in Tech Summit, 2018.

“Challenges of Child Maltreatment Prediction Models”, Allegheny County Department of Human Services, 2017.

“Leveraging Multidimensional Autocorrelations to Boost Sensitivity of Spectral Anomaly Detection”, DNDO Annual Academic Research Initiative Grantees Conference, 2015. Best Student Presentation Award.

“Author Profiling, an Application of Computational Linguistics”, Colombian Congress of Young Linguists, 2013.

Posters

Maria De­-Arteaga, Peter Huggins, Jonathan Elmer, Gilles Clermont, Artur Dubrawski, “Canonical Autocorrelation Embeddings for Comatose Patient Characterization”, Neural Information Processing Systems (NIPS), Women in Machine Learning Workshop, 2017.

Maria De-Arteaga, Artur Dubrawski, “Discovery of Complex Anomalous Patterns of Sexual Violence in El Salvador”, Data for Policy, 2016. 1st Place Innovation Award on Data Science.

Maria De­-Arteaga, Artur Dubrawski, Peter Huggins, “Canonical Autocorrelation Analysis for Radiation Threat Detection”, CRA­-W Grad Cohort Workshop, 2016.

William Herlands, Maria De-­Arteaga, Daniel Neill, and Artur Dubrawski, “Lass0: Sparse Non-­Convex Regression by Local Search”, Neural Information Processing Systems (NIPS), Optimization for Machine Learning Workshop, 2015.

Maria De­-Arteaga, Artur Dubrawski, Peter Huggins, “Canonical Autocorrelation Analysis for Radiation Threat Detection”, Neural Information Processing Systems (NIPS), Women in Machine Learning Workshop, 2015.

Maria De-­Arteaga, Sergio Jimenez, Julia Baquero, George Dueñas and Sergio Mancera, “Author Profiling Using Corpus Statistics, Lexicons and Stylistic Features” 10th PAN Evaluation Lab on Uncovering Plagiarism, Authorship. and Social Misuse, CLEF, 2013.