Albert Dorador

I am a PhD Candidate in the Department of Statistics at the University of Wisconsin - Madison, "la Caixa" Fellow. Google Scholar.

Research area | Interpretable Machine Learning (developing new algorithms and theory)

Currently working on TRUST (Transparent, Robust and Ultra-Sparse Trees), the most interpretable model tree algorithm ever created. Preliminary results (coming soon) show that it is at least as accurate as off-the-shelf state-of-the-art models, including deep neural networks, XGBoost, Random Forest, splines and RuleFit. See a recent live coding demo below where I show how Google's Gemini is integrated within my algorithm.

Co-advised by Professor Wei-Yin Loh and Emeritus Professor Zhengjun Zhang

Academic Background | M.S. Statistics & O.R. (Barcelona Tech), B.S. Business Economics and Bachelor of Laws (Universitat Pompeu Fabra), STEM exchange student-athlete (Rowing) at Carnegie Mellon University

Professional Background | Financial Risk Analyst (European Central Bank),  Transaction Advisory Services (Deloitte)

Contact | albert [dot] dorador [at] wisc [dot] edu

UW-Madison, Department of Statistics, B248 A, Medical Sciences Center

------------------------------------------------------------------------------------------------------------------

Current lead TA for ECE 761: Mathematical Machine Learning I (PhD course)

------------------------------------------------------------------------------------------------------------------

Sept. '23: Delighted to report that all my PhD students in STAT 610 have passed their Qualifying exam! Way to go!

------------------------------------------------------------------------------------------------------------------

Favorite quotes related to Interpretable ML

"Interpretable AI is an enhancement of human decision making, black box AI is a replacement of it." Afnan et al. '21

"Everything should be made as simple as possible, but not simpler." Albert Einstein

"If it doesn't have to produce correct results, I can make it arbitrarily fast." Gerald Weinberg (NASA, first man-in-space program)

"If it doesn't have to produce correct results, I can make it arbitrarily interpretable." 'Anonym'

Other favorite quotes related to academia (and beyond)

"Ever Tried. Ever failed. No matter. Try again. Fail again. Fail better." Samuel Beckett

"All models are wrong, but some are useful." George Box, founding Professor of our Statistics Department here at UW-Madison

"Per ardua ad astra". Latin saying

TRUST x Gemini