I am a PhD Candidate in the Department of Statistics at the University of Wisconsin - Madison, “la Caixa” Fellow, advised by Professor Wei-Yin Loh. My focus is on trustworthy AI, and I am currently developing TRUST (Transparent, Robust and Ultra-Sparse Trees) - the most interpretable model tree algorithm ever created. Preliminary results (coming soon) show that it often matches or exceeds the accuracy of leading black-box machine learning models like Random Forests, while remaining fully explainable. My goal is simple: to provide innovative and safe AI tools that allow users in high-stakes domains to stop choosing between accuracy and interpretability - and, in doing so, make a positive impact on society.

For a (literal) bird’s eye view of the places where I have worked, studied or given talks, check out this interactive map.

Research

Since 2022, my research has focused on trustworthy AI, first studying ensemble pruning as a way to improve both accuracy and interpretability of tree ensembles; then I developed my own model tree algorithm (TRUST), displaying excellent accuracy coupled with unparalleled interpretability. Prior work focused on quantitative finance, spanning both risk and portfolio management e.g. stop-loss rules.

Selected Publications

• Theoretical and Empirical Advances in Forest Pruning
  Dorador, A. Conference on Parsimony and Learning, Proceedings of Machine Learning Research (PMLR) (forthcoming), 2025.
  preprint

• On the efficacy of stop-loss rules in the presence of overnight gaps
  Arratia, A. and Dorador, A. Quantitative Finance, 19:11, 1857-1873, 2019.
  paper

Work Experience

University of Wisconsin - Madison

Ph.D. Researcher, Machine Learning (Sep 2022 - present)

• Doing research on tree-based methods with Professor Wei-Yin Loh, with focus on interpretability
• End-to-end development of my TRUST algorithm (Transparent, Robust and Ultra-Sparse Trees)
• Wrote a conference paper to appear in leading peer-reviewed machine learning journal [1]

Graduate Teaching Assistant (Sep 2021 - present)

• Machine Learning and Data Science courses: CS 861 (graduate), ECE 761 (graduate), STAT 340, STAT 333, STAT 240
• Probability and Mathematical Statistics courses: STAT 610 (graduate), STAT 311

European Central Bank

Risk Analyst, Directorate Risk Management, RAN division (Aug 2018 - Aug 2019)

• Statistical modeling of price reasonability tolerance bands
• Automation of core financial risk control tasks

Trainee, Directorate Risk Management, RAN division (Aug 2017 - Jul 2018)

• Led an innovative Machine Learning application in the bond market
• Executed core financial risk control tasks

Deloitte

Intern, Transaction Advisory Services (Sep 2014 - Dec 2014)

• Forensic and M&A / due dilligence rotations

Education

University of Wisconsin - Madison

Ph.D. in Statistics (Sep 2019 - present)

• Dissertation on Trustworthy AI (TRUST, ensemble pruning [1])
• Research visit at the Institute for Mathematical Sciences, National University of Singapore (Summer 2024)

Polytechnic University of Catalonia (BarcelonaTech)

M.S. in Statistics & Operations Research (Sep 2015 - Jun 2017)

• Thesis on quantitative financial risk management, deriving in a top peer-reviewed journal publication [2]

Carnegie Mellon University

Exchange student in the depts. of Mathematics, Computer Science and Business (Jan 2015 - May 2015)

• Carnegie Mellon Rowing Team (Tartan Crew)

Pompeu Fabra University

B.S. in Business Economics and Bachelor of Laws (Sep 2008 - Jun 2014)

• Honors double degree program, graduating 1st within the Business + Law cohort

Contact

If you would like to learn more about how my TRUST regression algorithm can help your business, reach out at trustalgorithm(dot)dev[at]gmail(dot)com. For other collaboration opportunities, please contact me at dorador(dot)albert[at]gmail(dot)com.