Antonio Padua Braga
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Antonio Padua Braga has a total of 165 co-authors. Here are the top co-authors by number of shared publications:
- Luiz Carlos Bambirra Torres
Department of Computing and Systems, Universidade Federal de Ouro Preto, João Monlevade, Brazil
10 shared publications - Frederico Coelho
UFMG
8 shared publications - Cristiano Leite Castro
Graduate Program in Electrical Engineering, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
6 shared publications - Marcelo Azevedo Costa
Universidade Federal de Minas Gerais
6 shared publications - Bruno Henrique Groenner Barbosa
Department of Automatics, Federal University of Lavras, Lavras, Minas Gerais, Brazil
4 shared publications Frederico Coelho
4 shared publications
Latest Publications
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (7) , 8535-8546
Locally Linear Continual Learning for Time Series Based on VC-Theoretical Generalization Bounds
Most machine learning methods assume fixed probability distributions, limiting their applicability in nonstationary real-world scenarios. While continual learning methods address this issue, current approaches often rely on closed-box models or require extensive user intervention for interpretability. We propose SyMPLER (Systems Modeling through Piecewise Linear Evolving Regression), an explainable model for time series forecasting in nonstationary environments based on dynamic piecewise-linear approximations. Unlike other locally linear models, SyMPLER uses generalization bounds from Statistical Learning Theory to automatically determine when to add new local models based on prediction errors, eliminating the need for explicit clustering of the data. Experiments show that SyMPLER can achieve comparable performance to both closed-box and existing explainable models while maintaining a human-interpretable structure that reveals insights about the system's behavior. In this sense, our approach conciliates accuracy and interpretability, offering a transparent and adaptive solution for forecasting nonstationary time series.
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (7) , 8535-8546
Locally Linear Continual Learning for Time Series Based on VC-Theoretical Generalization Bounds
Most machine learning methods assume fixed probability distributions, limiting their applicability in nonstationary real-world scenarios. While continual learning methods address this issue, current approaches often rely on closed-box models or require extensive user intervention for interpretability. We propose SyMPLER (Systems Modeling through Piecewise Linear Evolving Regression), an explainable model for time series forecasting in nonstationary environments based on dynamic piecewise-linear approximations. Unlike other locally linear models, SyMPLER uses generalization bounds from Statistical Learning Theory to automatically determine when to add new local models based on prediction errors, eliminating the need for explicit clustering of the data. Experiments show that SyMPLER can achieve comparable performance to both closed-box and existing explainable models while maintaining a human-interpretable structure that reveals insights about the system's behavior. In this sense, our approach conciliates accuracy and interpretability, offering a transparent and adaptive solution for forecasting nonstationary time series.
- Open Access
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 41633-41649
A Deep Reinforcement Learning Approach for Fault Detection in Continuous Chemical Processes
Fault detection in continuous industrial processes is crucial to ensure production, quality, safety, environmental and economic aspects. Reliable fault detection systems have become even more necessary due to increasingly stringent regulations. The complexity of processes, on the one hand, and the availability of data, on the other, have spread the adoption of the data-driven approach for industrial fault detection. Deep Reinforcement Learning (RL) has recently been applied for this purpose, mainly in rotating machinery and rolling bearings. In chemical processes, most RL applications are in process control. The main contribution of this work lies in proposing a reinforcement learning (RL)-based framework for fault detection in continuous chemical processes. The challenges are related to the large number of variables, nonlinear relationships and changing operational dynamics, to name a few. The classic Tennessee Eastman Process (TEP) benchmark was used as a case study. The results were compared with Principal Component Analysis (PCA), commonly used for chemical process monitoring, dynamic PCA (DPCA), and neural networks. Compared to PCA, the RL approach showed considerably lower missed detection rate (MDR) and time to detection (TTD) while maintaining reasonable false alarm rate (FAR) for all hard-to-detect faults, contributing to early fault detectability. Similar results were obtained with regard to DPCA. Compared to the supervised approach, RL proved to be more flexible regarding model parameterization, which can be very useful for practical aspects of industrial fault detection. The results show that RL-based approaches can be very useful for fault detection in continuous, non-linear, and time-varying systems, such as chemical processes. For example, with regard to adaptive alarm thresholds and the detection of known and unknown faults. The latter represents a major challenge in industrial practice.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 41633-41649
Open AccessA Deep Reinforcement Learning Approach for Fault Detection in Continuous Chemical Processes
Fault detection in continuous industrial processes is crucial to ensure production, quality, safety, environmental and economic aspects. Reliable fault detection systems have become even more necessary due to increasingly stringent regulations. The complexity of processes, on the one hand, and the availability of data, on the other, have spread the adoption of the data-driven approach for industrial fault detection. Deep Reinforcement Learning (RL) has recently been applied for this purpose, mainly in rotating machinery and rolling bearings. In chemical processes, most RL applications are in process control. The main contribution of this work lies in proposing a reinforcement learning (RL)-based framework for fault detection in continuous chemical processes. The challenges are related to the large number of variables, nonlinear relationships and changing operational dynamics, to name a few. The classic Tennessee Eastman Process (TEP) benchmark was used as a case study. The results were compared with Principal Component Analysis (PCA), commonly used for chemical process monitoring, dynamic PCA (DPCA), and neural networks. Compared to PCA, the RL approach showed considerably lower missed detection rate (MDR) and time to detection (TTD) while maintaining reasonable false alarm rate (FAR) for all hard-to-detect faults, contributing to early fault detectability. Similar results were obtained with regard to DPCA. Compared to the supervised approach, RL proved to be more flexible regarding model parameterization, which can be very useful for practical aspects of industrial fault detection. The results show that RL-based approaches can be very useful for fault detection in continuous, non-linear, and time-varying systems, such as chemical processes. For example, with regard to adaptive alarm thresholds and the detection of known and unknown faults. The latter represents a major challenge in industrial practice.
- 5 November 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 186
Random projection large margin algorithm for binary classification problems
- 5 November 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 186
Random projection large margin algorithm for binary classification problems
- 25 September 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 198, 43-49
Large margin classifier with graph-based adaptive regularization
- 25 September 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 198, 43-49
Large margin classifier with graph-based adaptive regularization
- Open Access
- 30 July 2025
- Published by Universidade Federal de Minas Gerais - Pro-Reitoria de Pesquisa in Revista Docência do Ensino Superior
- Vol. 15, 1-16
Entrevista com a Comissão Permanente de Inteligência Artificial da UFMG
Esta entrevista, realizada com a Comissão Permanente de Inteligência Artificial em maio de 2025, em ambiente virtual compartilhado, tem como objetivo compartilhar as ações já executadas e as atividades planejadas pela Comissão, além de discutir aspectos de sua criação no âmbito do ensino superior. A Comissão é organizada em coordenações, e as perguntas foram formuladas seguindo essa estrutura interna, contando com a participação de 12 professores membros, que discorrem sobre a inteligência artificial no contexto do ensino, da pesquisa, da extensão e da administração – pilares que sustentam a universidade e que são diretamente impactados pelo uso dessas tecnologias. O texto apresenta a Comissão à comunidade acadêmica e enseja um exercício de reflexão sobre o impacto da inteligência artificial no ensino superior no intuito de contribuir para a consolidação de uma sociedade mais informada, letrada e capacitada para lidar com as transformações decorrentes das tecnologias emergentes.
- 30 July 2025
- Published by Universidade Federal de Minas Gerais - Pro-Reitoria de Pesquisa in Revista Docência do Ensino Superior
- Vol. 15, 1-16
Open AccessEntrevista com a Comissão Permanente de Inteligência Artificial da UFMG
Esta entrevista, realizada com a Comissão Permanente de Inteligência Artificial em maio de 2025, em ambiente virtual compartilhado, tem como objetivo compartilhar as ações já executadas e as atividades planejadas pela Comissão, além de discutir aspectos de sua criação no âmbito do ensino superior. A Comissão é organizada em coordenações, e as perguntas foram formuladas seguindo essa estrutura interna, contando com a participação de 12 professores membros, que discorrem sobre a inteligência artificial no contexto do ensino, da pesquisa, da extensão e da administração – pilares que sustentam a universidade e que são diretamente impactados pelo uso dessas tecnologias. O texto apresenta a Comissão à comunidade acadêmica e enseja um exercício de reflexão sobre o impacto da inteligência artificial no ensino superior no intuito de contribuir para a consolidação de uma sociedade mais informada, letrada e capacitada para lidar com as transformações decorrentes das tecnologias emergentes.
- 4 January 2025
- Published by Elsevier in Pattern Recognition
- Vol. 162
A projected gradient solution to the minimum connector problem with extensions to support vector machines
- 4 January 2025
- Published by Elsevier in Pattern Recognition
- Vol. 162
A projected gradient solution to the minimum connector problem with extensions to support vector machines
- 18 December 2024
- Published by Elsevier in Information Sciences
- Vol. 700
Graph-based method for autonomous adaptation in online learning of non-stationary data
- 18 December 2024
- Published by Elsevier in Information Sciences
- Vol. 700
Graph-based method for autonomous adaptation in online learning of non-stationary data
- 31 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (5) , 8307-8316
Multiclass Graph-Based Large Margin Classifiers: Unified Approach for Support Vectors and Neural Networks
While large margin classifiers are originally an outcome of an optimization framework, support vectors (SVs) can be obtained from geometric approaches. This article presents advances in the use of Gabriel graphs (GGs) in binary and mul-ticlass classification problems. For Chipclass, a hyperparameter-less and optimization-less GG-based binary classifier, we discuss how activation functions and support edge (SE)-centered neurons affect the classification, proposing smoother functions and structural SV (SSV)-centered neurons to achieve margins with low probabilities and smoother classification contours We extend the neural network architecture, which can be trained with backpropagation with a softmax function and a cross-entropy loss, or by solving a system of linear equations. A new subgraph-/distance-based membership function for graph regularization is also proposed, along with a new GG recomputation algorithm that is less computationally expensive than the standard approach. Experimental results with the Friedman test show that our method was better than previous GG-based classifiers and statistically equivalent to tree-based models.
- 31 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (5) , 8307-8316
Multiclass Graph-Based Large Margin Classifiers: Unified Approach for Support Vectors and Neural Networks
While large margin classifiers are originally an outcome of an optimization framework, support vectors (SVs) can be obtained from geometric approaches. This article presents advances in the use of Gabriel graphs (GGs) in binary and mul-ticlass classification problems. For Chipclass, a hyperparameter-less and optimization-less GG-based binary classifier, we discuss how activation functions and support edge (SE)-centered neurons affect the classification, proposing smoother functions and structural SV (SSV)-centered neurons to achieve margins with low probabilities and smoother classification contours We extend the neural network architecture, which can be trained with backpropagation with a softmax function and a cross-entropy loss, or by solving a system of linear equations. A new subgraph-/distance-based membership function for graph regularization is also proposed, along with a new GG recomputation algorithm that is less computationally expensive than the standard approach. Experimental results with the Friedman test show that our method was better than previous GG-based classifiers and statistically equivalent to tree-based models.
- Open Access
- 18 October 2023
- Published by SWGE Sistemas Ltda in Anais Do Vi Simpósio Brasileiro de Sistemas Elétricos
- Vol. 1 (2)
Graph-based Clustering Index for Topology and Classification Analysis
This work proposes a quality index based on a graph of distance relationships, the Gabriel Graph, to analyze the performance of classification models. The results have shown a consistent correlation between the quality index and the model’s predictions, both at an individual and global level. These findings emphasize the importance of considering the data space to explain and optimize the performance of classification models.
- 18 October 2023
- Published by SWGE Sistemas Ltda in Anais Do Vi Simpósio Brasileiro de Sistemas Elétricos
- Vol. 1 (2)
Open AccessGraph-based Clustering Index for Topology and Classification Analysis
This work proposes a quality index based on a graph of distance relationships, the Gabriel Graph, to analyze the performance of classification models. The results have shown a consistent correlation between the quality index and the model’s predictions, both at an individual and global level. These findings emphasize the importance of considering the data space to explain and optimize the performance of classification models.
- Open Access
- 3 May 2023
- Published by Editora UNIVALI
- Vol. 14, 202-208
Algoritmo de Treinamento para uma Rede SLFN com Projeção Aleatória e Margem Larga Learning Algorithm for an SLFN Network with Random Projection and Large Margin
This work presents a large margin learning algorithm for single hidden layer feedforward networks (SLFNs) with random weights for the hidden neurons, called RP-IMA. This algorithm, applied to binary classification problems, proposes randomly assignedweights to the hidden layer and the use of an incremental margin algorithm to calculate the weights of the output neuron of the SLFN. The results showed that the proposed algorithm is capable to obtain a large separation margin in the feature space and has its performance less sensitive to variations in the network architecture, when compared to Extreme Learning Machines.
- 3 May 2023
- Published by Editora UNIVALI
- Vol. 14, 202-208
Open AccessAlgoritmo de Treinamento para uma Rede SLFN com Projeção Aleatória e Margem Larga Learning Algorithm for an SLFN Network with Random Projection and Large Margin
This work presents a large margin learning algorithm for single hidden layer feedforward networks (SLFNs) with random weights for the hidden neurons, called RP-IMA. This algorithm, applied to binary classification problems, proposes randomly assignedweights to the hidden layer and the use of an incremental margin algorithm to calculate the weights of the output neuron of the SLFN. The results showed that the proposed algorithm is capable to obtain a large separation margin in the feature space and has its performance less sensitive to variations in the network architecture, when compared to Extreme Learning Machines.