Research
My research focuses on the methodology and theory of machine learning and statistics, with a flair of nonparametrics. I also do applications, usually in the form of collaborations.
Methodology and Theory
Prediction-Powered E-Values (preprint)
Daniel Csillag, Claudio José Struchiner, Guilherme Tegoni Goedert
preprint
Image Super-Resolution with Guarantees via Conformal Generative Models (preprint)
Eduardo Adame, Daniel Csillag, Guilherme Tegoni Goedert
preprint
Strategic Conformal Prediction
Daniel Csillag, Claudio José Struchiner, Guilherme Tegoni Goedert
AISTATS 2025
Diffusion-Powered Image Super-Resolution that you can Actually Trust (workshop)
Daniel Csillag, Eduardo Adame, Guilherme Tegoni Goedert
NeurIPS 2024 workshop on Statistical Frontiers in LLMs and Foundation Models
Generalization Bounds for Causal Regression: Insights, Guarantees and Sensitivity Analysis
Daniel Csillag, Claudio J. Struchiner, Guilherme T. Goedert
ICML 2024
ExactBoost: Directly Boosting the Margin in Combinatorial and Non-decomposable Metrics
Daniel Csillag, Carolina Piazza, Thiago Ramos, João Vitor Romano, Roberto Imbuzeiro Oliveira, Paulo Orenstein
AISTATS 2022
Optimizing Combinatorial and Non-decomposable Metrics with ExactBoost (workshop)
Daniel Csillag, Carolina Piazza, Thiago Ramos, João Vitor Romano, Roberto Imbuzeiro Oliveira, Paulo Orenstein
ICML 2021 workshop on Beyond First Order Methods in Machine Learning Systems (OPTICML2021)
Applications in Epidemiology and Medicine
The impact of vaccination on the length of stay of hospitalized COVID-19 patients in Brazil
Cleber Vinicius Brito dos Santos, Lara Esteves Coelho, Tatiana Guimarães de Noronha, Guilherme Tegoni Goedert, Daniel Csillag, Paula Mendes Luz, Guilherme Loureiro Werneck, Daniel Antunes Maciel Villela, Claudio José Struchiner
Vaccine
How to best predict who experiences food insecurity using data from a population-based survey: implementing machine learning algorithms in epidemiological studies
Lara E. Coelho, Daniel Csillag, Guilherme T. Goedert, Débora C. Pires, Emilia M. Jalil, Hugo Perazzo, Thiago S. Torres, Sandra W. Cardoso, Eduardo M. Peixoto, Breno Augusto Bormann de Souza Filho, Ana T. R. Vasconscelos, Guilherme C. da Fonseca, Liliane T. F. Cavalcante, Carlos A. M. Costa, Rodrigo T. Amancio, Sandro Nazer, Daniel A. M. Villela, Tiago Pereira, Cleber V. B. D. Santos, Nadia C. P. Rodrigues, Mariângela Freitas Silveira, Beatriz Grinsztejn, Valdilea G. Veloso, Paula M. Luz, Claudio J. Struchiner
Under review
SARS-CoV-2 transmission in a highly vulnerable population of Brazil: a household cohort study
Lara E. Coelho, Paula M. Luz, Débora C. Pires, Emilia M. Jalil, Hugo Perazzo, Thiago S. Torres, Sandra W. Cardoso, Eduardo M. Peixoto, Sandro Nazer, Eduardo Massad, Luiz Max Fagundes de Carvalho, Weeberb J. Réquia, Fernando do Couto Motta, Marilda Mendonça Siqueira, Ana Tereza R. Vasconcelos, Guilherme C. da Fonseca, Liliane T. F. Cavalcante, Carlos A. M. Costa, Rodrigo T. Amancio, Daniel A. M. Villela, Tiago Pereira, Guilherme T. Goedert, Cleber V. B. D. Santos, Nadia C. P. Rodrigues, Breno Augusto Bormann, Daniel Csillag, Beatriz Grinsztejn, Valdilea G. Veloso, Claudio J. Struchiner
The Lancet Regional Health -- Americas
AmnioML: Amniotic Fluid Segmentation and Volume Prediction with Uncertainty Quantification
Daniel Csillag, Lucas Monteiro, Thiago Ramos, João Vitor Romano, Rodrigo Schuller, Roberto Beauclair Seixas, Roberto Imbuzeiro Oliveira, Paulo Orenstein
AAAI-23 Special Programs, IAAI
Uncertainty Quantification for Amniotic Fluid Segmentation and Volume Prediction (workshop)
Daniel Csillag, Lucas Monteiro, Thiago Ramos, João Vitor Romano, Rodrigo Schuller, Roberto Beauclair Seixas, Roberto Imbuzeiro Oliveira, Paulo Orenstein
ICML2021 workshop on Interpretable Machine Learning in Healthcare (ICLH2021)