Updates

I hope now that I have a bit more time, I can write more here. In the last few months, I concluded my Ph.

Plot Panels in R (from brms models)

This week I had to make a panel with results from 5 different brms models in R. The ggplot2 allows you to do facet_wraps to create a panel with multiple plots but it also requires that you have a well formated dataframe.

Paper accepted at IEEE Transactions on Evolutionary Computation!

I am happy to announce that our paper “Statistical Models for the Analysis of Optimization Algorithms with Benchmark Functions” by David Issa Mattos, Jan Bosch and Helena Holmström Olsson has been accepted for publication at the IEEE Transactions on Evolutionary Computation (DOI:10.

Structuring a Latex project for a paper

Over the last five years, I have written papers in different combinations of Word + Mendeley/Zotero, Latex/Overleaf, Google Docs + Zotero, Word Online + Mendeley.

On the Assessment of Benchmark Suites for Algorithm Comparison

Today, we submitted a new paper (On the Assessment of Benchmark Suites for Algorithm Comparison) to the IEEE Transactions on Evolutionary Computing (and of course to Arxiv).

bpcs - release v.1.2.0

We have a new release of the bpcs package on Github. This release introduces a few new features, removed dependencies and mostly moved from rstan to cmdstanr.

Writing R packages With Stan

Now, I have been developing the bpcs package for the last 4 months I thought of sharing a bit my experience with developing a R package with Stan.

SE Seminar 2021 - “Bayesian Bradley-Terry models: from primates to machine learning”

This Feb 5th, I will present some of my research around Bradley-Terry models in the SE Seminar at Chalmers. My goal is to show a the generality of the Bayesian BT model and that it can be used for both behavior research as well as software engineering.

Bayesian Paired-Comparison with the bpcs Package

Yesterday, we submitted the paper: Mattos, D. I. and Ramos, E. M. S. “Bayesian Paired-Comparison with the bpcs Package” arxiv:2101.11227

bpcs - release v.1.1

A new release of the bpcs package is available in Github. This release introduces new models and correct some bugs. This release will be submitted to CRAN (not there yet) and the on-going paper we are writing is based on this release.