Bruna Wundervald

Bruna Wundervald
National University of Ireland, Maynooth | NUI Maynooth · Hamilton Institute

PhD Candidate in Bayesian Machine Learning

About

25
Publications
13,698
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
14
Citations

Publications

Publications (25)
Preprint
Full-text available
We propose a simple yet powerful extension of Bayesian Additive Regression Trees which we name Hierarchical Embedded BART (HE-BART). The model allows for random effects to be included at the terminal node level of a set of regression trees, making HE-BART a non-parametric alternative to mixed effects models which avoids the need for the user to spe...
Article
Music is an important part of most people’s lives and also of the culture of a country. Moreover, the different characteristics of songs, such as genre and the chord sequences, could have different impacts on individual behaviours. Even considering just seven chords and the respective variations, originality can be a crucial element of a song’s suc...
Article
Full-text available
Increasingly, music recommendations are influencing user listening behavior, which naturally impacts the music industry, as well as the cultural and social aspects of our society. This has opened up a research area, namely the identification of biases introduced by recommender systems in the music context. Recent research, which focused on users of...
Article
Full-text available
We develop a new approach for feature selection via gain penalization in tree-based models. First, we show that previous methods do not perform sufficient regularization and often exhibit sub-optimal out-of-sample performance, especially when correlated features are present. Instead, we develop a new gain penalization idea that exhibits a general l...
Conference Paper
Full-text available
Many factors are involved in the definition of music genres, making it an active area of research. This work focuses on verifying the connection between harmonic information and genre specification in Brazilian music, through the evaluation of feature importance in machine learning models. We construct four different sets of manually engineered har...
Preprint
Full-text available
We develop a new approach for feature selection via gain penalization in tree-based models. First, we show that previous methods do not perform sufficient regularization and often exhibit sub-optimal out-of-sample performance, especially when correlated features are present. Instead, we develop a new gain penalization idea that exhibits a general l...
Presentation
Full-text available
This talk-workshop is suited to all levels of R. You can expect from it: 1. An overview of the theory of tree-based models: decision trees and random forests; 2. Learn how to implement and interpret random forests in R
Technical Report
Full-text available
Bayesian methods are an alternative to standard frequentist methods and as a result, have gained popularity. This report will display some of the fundamental ideas in Bayesian modeling and will present both the theory behind Bayesian statistics and some practical examples of Bayesian linear regression. Simulated data and real-world data were used t...
Preprint
Full-text available
Music genre can be hard to describe: many factors are involved, such as style, music technique, and historical context. Some genres even have overlapping characteristics. Looking for a better understanding of how music genres are related to musical harmonic structures, we gathered data about the music chords for thousands of popular Brazilian songs...
Technical Report
Full-text available
Presentation
Full-text available
Presentation
Full-text available

Network

Cited By