Machine Learning

NLS: An accurate and yet easy-to-interpret prediction method
Over the last years, the predictive power of supervised machine learning (ML) has undergone impressive advances, achieving the status …
V. Coscrato, M. H. A. Inacio, T. Botari, Rafael Izbicki
January, 2023 Neural Networks
Preprint PDF
A new LDA formulation with covariates
G. Shimizu, Rafael Izbicki, D. Valle
January, 2023 Communications in Statistics - Simulation and Computation
Preprint
CD-split and HPD-split: Efficient conformal regions in high dimensions
Conformal methods create prediction bands that control average coverage assuming solely i.i.d. data. We introduce CD-split and HPD-split, which yield general prediction regions and converge to the optimal highest predictive density set.
Rafael Izbicki, Gilson Shimizu, Rafael B. Stern
May, 2022 Journal of Machine Learning Research
PDF Cite Code
CD-split and HPD-split: Efficient conformal regions in high dimensions
Likelihood-Free Frequentist Inference: Confidence Sets with Correct Conditional Coverage
Many areas of science make extensive use of computer simulators that implicitly encode likelihood functions of complex systems. …
N. Dalmasso, L. Masserano, D. Zhao, Rafael Izbicki, A. B. Lee
April, 2022 Submitted
PDF
Causas de desaparecimento no estado de São Paulo entre 2013 e 2014: uma análise automatizada de boletins de ocorrência
Este artigo estuda as causas de desaparecimento no estado de São Paulo por idade e sexo dos desaparecidos. Para tanto, utiliza algoritmos de aprendizado de máquina para avaliar automaticamente o texto de boletins de ocorrência.
Mateus B. Comito, Rafael Izbicki, Rafael B. Stern, Julio A. Z. Trecenti
February, 2021 In RDPESP
PDF Cite
Identifying Distributional Differences in Convective Evolution Prior to Rapid Intensification in Tropical Cyclones
Tropical cyclone (TC) intensity forecasts are issued by human forecasters who evaluate spatio-temporal observations (e.g., satellite …
T. McNeely, G. Vincent, Rafael Izbicki, K. M. Wood, A. B. Lee
February, 2021 Tackling Climate Change with Machine Learning workshop at NeurIPS 2021 (poster)
PDF
The Latent Dirichlet Allocation model with covariates (LDAcov): a case study on the effect of fire on species composition in Amazonian forests
Understanding and predicting the effect of global change phenomena on biodiversity is challenging given that biodiversity data are …
D. Valle, G. Shimizu, Rafael Izbicki, L. Maracahipes, D. Silvério, L. Paolucci, Y. Jameel, P. Brando
February, 2021 Ecology and Evolution
PDF
Distance assessment and analysis of high-dimensional samples using variational autoencoders
An important question in many machine learning applications is whether two samples arise from the same generating distribution. …
M. H. de A. Inácio, Rafael Izbicki, B. Gyires-Tóth
January, 2021 Information Sciences
Preprint PDF
Flexible distribution-free conditional predictive bands using density estimators
Conformal methods create prediction bands that control average coverage assuming solely i.i.d. data. Besides average coverage, one …
Gilson Shimizu, Rafael Izbicki, Rafael B. Stern
April, 2020 In PMLR
PDF Cite Source Document
Covid-19 Einstein Analysis
In this post I analyse the covid-19 data from https://www.kaggle.com/einsteindata4u/covid19, which contains information about patients from Albert Einstein’s Hospital, in São Paulo (Brazil). My main assumptions in the following analysis are that:
Mar 29, 2020 8 min read R, machine learning, Blog Post