activity
20242026
collaborators

6 papers

stat.ML2026

LOCUS: A Distribution-Free Loss-Quantile Score for Risk-Aware Predictions

Matheus Barreto, Mário de Castro, Thiago R. Ramos +2

Modern machine learning models can be accurate on average yet still make mistakes that dominate deployment cost. We introduce Locus, a distribution-free wrapper that produces a per…

cs.DB2025

From Patents to Dataset: Scraping for Oxide Glass Compositions and Properties

Gustavo Laranja Thomaello, Thomaz Yeiden Busnardo Aguena, Eric Trevelato Costa +4

In this work, we present web scraping techniques to extract in- formation from patent tables, clean and structure them for future use in predictive machine learning models to devel…

stat.ML2025

Conformal Prediction for Compositional Data

Lucas P. Amaral, Luben M. C. Cabezas, Thiago R. Ramos +1

Dirichlet regression models are suitable for compositional data, in which the response variable represents proportions that sum to one. However, there are still no well-established…

stat.ML2025

Epistemic Uncertainty in Conformal Scores: A Unified Approach

Luben M. C. Cabezas, Vagner S. Santos, Thiago R. Ramos +1

Conformal prediction methods create prediction bands with distribution-free guarantees but do not explicitly capture epistemic uncertainty, which can lead to overconfident predicti…

stat.ME2024

Conformal Calibration of Statistical Confidence Sets

Luben M. C. Cabezas, Guilherme P. Soares, Thiago R. Ramos +2

Constructing valid confidence sets is a crucial task in statistical inference, yet traditional methods often face challenges when dealing with complex models or limited observed sa…

cs.LG2024

PersonalizedUS: Interpretable Breast Cancer Risk Assessment with Local Coverage Uncertainty Quantification

Alek Fröhlich, Thiago Ramos, Gustavo Cabello +3

Correctly assessing the malignancy of breast lesions identified during ultrasound examinations is crucial for effective clinical decision-making. However, the current "golden stand…