The main objective of this research is the conception of an approach to predict the risk of morta... more The main objective of this research is the conception of an approach to predict the risk of mortality in ICUs. A cohort of 17,734 patients was used, from the MIMIC-III Database, considering 10 input predictor variables and 8 Machine Learning methods. The best performance was achieved by the Gradient Boosting Machine (GBM) method, which obtained 0,843 (±0,015) of AUC and 0,503 (±0,048) of F1 score. The results are promising and, in some cases, superior to those obtained by other proposals identified in the literature review. Resumo. Esta pesquisa tem por objetivo central a concepção de uma abordagem para predição do risco de mortalidade em UTIs. Foi empregada uma coorte de 17.734 pacientes provenientes do Banco de Dados MIMIC-III, sendo consideradas 10 variáveis preditoras de entrada e 8 métodos de Aprendizado de Máquina. A melhor performance foi alcançada pelo método Gradient Boosting Machine (GBM), que atingiu 0,843 (±0,015) de AUC e 0,503 (±0,048) de F1 score. Os resultados são promissores e, em alguns casos, superiores a outras propostas identificadas na revisão de literatura.
2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Jul 18, 2022
This paper considers the consensual analysis in decision-making (CDM) processes based on fuzzy lo... more This paper considers the consensual analysis in decision-making (CDM) processes based on fuzzy logic (FL) and interval-valued fuzzy logic (IVFL), providing a CDM-strategy, by exploring the axiomatic properties of fuzzy consensus measures (FCM) via penalty functions. Thus, two models are formalized, FS-FCM and IVFS-FCM. In the former, the fuzzy-valued lattice enables the analysis of fuzzy information for linguistic variables (LV), which is obtained by the aggregation of penalty functions. And, in the latter, the consensus measures of fuzzy sets are aggregated to build a new consensual analysis modeling. Thus, e.g., the cohesion of several terms related to the same LV can be analyzed, and also the coherence between fuzzy sets referring to the lowest and highest projections. Such models decide based on relevance criteria and qualitative assessments, via the selection of alternatives, supporting the corresponding algorithmic strategies: FS-FCM strategy, applied to fuzzy values, and IVFS-FCM strategy, covering fuzzy sets. The Intf-HybridMem approach explores the access patterns to volatile and non-volatile memories related to decision-making in two steps: (i) the FS-FCM strategy explores consensus measures of fuzzy values from membership functions; and (ii) the IVFS-FCM strategy, modeling inaccuracy inherent in input variables, as read/write frequency and access recency, also including the migration recommendation as output, which is validated by evaluations carried out in both proposed strategies.
Playful and unplugged activities have been proposed as an improved alternative enabling the inser... more Playful and unplugged activities have been proposed as an improved alternative enabling the insertion of methodological innovation to increase learning in elementary education, through the application of computational concepts and techniques. In such context, the consolidation of the Space Adventure activity is described, presenting the modelling of its final stage named as Task 5. In this stage, the game execution promotes learning via simulation of a scientific space journey. The concepts of abstraction, decomposition, pattern recognition are worked on, and the search for solutions for energy efficiency and correction evaluation of data organization is also encouraged. Resumo. Atividades lúdicas e desplugadas têm sido propostas como alternativas viáveis economicamente e com inserção de inovação metodológica para incremento da aprendizagem na Educação Fundamental, via aplicação de conceitos e técnicas da computação. Neste contexto, descreve-se a consolidação da atividade Aventura Espacial, apresentando a modelagem da sua etapa final, identificada como Tarefa 5. Nesta etapa, a execução das jogadas promove aprendizagem via simulação de uma jornada científica espacial. São trabalhados os conceitos de abstração, decomposição, reconhecimento de padrões e estimulam-se ainda a busca de soluções para eficiência energética e avaliação da correção na organização de dados.
Trends in Applied and Computational Mathematics, Oct 13, 2004
This paper introduces the stochastic version of the Geometric Machine Model for the modelling of ... more This paper introduces the stochastic version of the Geometric Machine Model for the modelling of sequential, alternative, parallel (synchronous) and nondeterministic computations with stochastic numbers stored in a (possibly infinite) shared memory. The programming language L(D → ∞), induced by the Coherence Space of Processes D → ∞ , can be applied to sequential and parallel products in order to provide recursive definitions for such processes, together with a domain-theoretic semantics of the Stochastic Arithmetic. We analyze both the spacial (ordinal) recursion, related to spacial modelling of the stochastic memory, and the temporal (structural) recursion, given by the inclusion relation modelling partial objects in the ordered structure of process construction.
The aim of this paper is to introduce the notion of interval additive generators of interval t-no... more The aim of this paper is to introduce the notion of interval additive generators of interval t-norms as interval representations of additive generators of t-norms, considering both the correctness and the optimality criteria, in order to provide a more systematic methodology for the selection of interval t-norms in the various applications. We prove that interval additive generators satisfy the main
2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Jul 18, 2022
Firstly, this work discusses the main conditions guarantying that general overlap (grouping) func... more Firstly, this work discusses the main conditions guarantying that general overlap (grouping) functions can be obtained from n-dimensional overlap (grouping) functions. Focusing on QL-implications, which are usually generated by strong negations together with t-norms and t-conorms, we consider a non-restrictive construction, by relaxing not only the associativity and the corresponding neutral elements (NE) but also the reverse construction of other properties. Thus, the main properties of the QL-implication class are studied, considering a tuple (G,N,O) generated from grouping and overlap functions together with the greatest fuzzy negation. In addition, in order to provide more flexibility, we define a subclass of QL-implications generated from general overlap and general grouping functions. Some examples are introduced, illustrating the constructive methods to generate such operators.
This paper explores consensus measures to support to determining the level of use of physical mac... more This paper explores consensus measures to support to determining the level of use of physical machines in a cloud computing environment. The fuzzy approach considers the uncertainties present in the cloud computing environment and the consensus analysis of the fuzzy sets is based on arithmetic and exponencial means. In the proposal evaluation, a case study aimed at the architecture of the Framework Int-FLBCC, which is under development by the research group was designed.
A main obstacle in simulation of quantum algorithms is the exponential increase in the temporal a... more A main obstacle in simulation of quantum algorithms is the exponential increase in the temporal and spatial complexities, especially in dense quantum transformations such as the Hadamard operator. In this work, new optimizations for the execution of quantum transformations in the Distributed Geometric Machine (D-GM) environment.Instead of executing them in a single step, they are decomposed and only values different from Identity operator are stored. As a benchmark, Hadamard Transformations were simulated up to 28 qubits in a GPU. When compared to our previous implementation, our new approach is 10, 829⇥ faster and allows for the simulation of more qubits. Resumo. Um dos maiores obstáculos para a simulação de algoritmos
Computer systems based on intuitionistic fuzzy logic are capable of generating a reliable output ... more Computer systems based on intuitionistic fuzzy logic are capable of generating a reliable output even when handling inaccurate input data by applying a rule based system, even with rules that are generated with imprecision. The main contribution of this paper is to show that quantum computing can be used to extend the class of intuitionistic fuzzy sets with respect to representing intuitionistic fuzzy Xor operators. This paper describes a multi-dimensional quantum register using aggregations operators such as t-(co)norms based on quantum gates allowing the modeling and interpretation of intuitionistic fuzzy Xor operations.
International Journal of High Performance Computing Applications, Jan 16, 2019
Exponential increase and global access to read/write memory states in quantum computing (QC) simu... more Exponential increase and global access to read/write memory states in quantum computing (QC) simulation limit both the number of qubits and quantum transformations which can be currently simulated. Although QC simulation is parallel by nature, spatial and temporal complexity are major performance hazards, making this a nontrivial application for high performance computing. A new methodology employing reduction and decomposition optimizations has shown relevant results, but its GPU implementation could be further improved. In this work, we develop a new kernel for in situ GPU simulation that better explores its resources without requiring further hardware. Shor's and Grover's algorithms are simulated up to 25 and 21 qubits respectively and compared to our previous version, to LIQUiji 0 s simulator and to ProjectQ framework, showing better results with relative speedups up to 4.38Â, 3357.76Â and 333Â respectively.
This paper presents seven concurrent hash table implementations in Haskell, ranging from low-leve... more This paper presents seven concurrent hash table implementations in Haskell, ranging from low-level synchronization mechanisms to high-level ones such as transactional memories. The hash tables were compared using different initial sizes, load factors, data types and hash functions. We also present a case study on implementing a color palette algorithm using the hash tables. The result of the comparison between the algorithms shows that the implementation using the STM Haskell transactional memory library and fine-grain synchronization provides the best performance and good scalability, followed by the implementation using lock striping and MVars.
Sustainable Computing: Informatics and Systems, Mar 1, 2021
One of the great challenges of today's computer memory architectures has been their energy consum... more One of the great challenges of today's computer memory architectures has been their energy consumption. In this sense, DRAM technology has come close to its scalability limit due to energy issues. Thus, new technologies have emerged intending to replace DRAM as main memory. In this context, non-volatile memories emerge promising to be an alternative for low energy consumption memory systems. Due to the challenges of these new technologies, hybrid approaches have been a trend. When working with hybrid memories, there is a necessity for decisions in which memory modules each data should be stored. This decision must consider the data access profile and the characteristics of memory technology. Thus, this work presents the Intf-HybridMem architecture, a proposal for page migration in hybrid memories using fuzzy systems to support decision making. The fuzzy approach can model the uncertainties of the data access profile and the characteristics of the memory modules. The tests seek to identify the performance limit of the Intf-HybridMem through the use of an oracle mechanism. Results show that there is potential for exploring migration mechanisms with overhead of 0.18% and 0.39% in some scenarios.
One of the main obstacles for the adoption of quantum algorithm simulation is the exponential inc... more One of the main obstacles for the adoption of quantum algorithm simulation is the exponential increase in temporal and spatial complexities, due to the expansion of transformations and read/write memory states by using tensor product in multi-dimension applications. Reduction and decomposition optimizations via the Id-operator provide a smart and appropriate storage and distribution of quantum information. Reductions are achieved by avoiding replication and sparsity inherited from such operators. By using decompositions, applications may be divided into sub-steps to store only distinct values from Id-operators, instead of executing quantum transformations in a single step. Additional optimizations based on mixed partial processes provide control over increase in read/write memory states in quantum transformations, also contributing to increase the scalability regarding hardware-GPUs memory limit. Hadamard and Discret Quantum Fourier Transforms were simulated up to 28 qubits applications over a single GPU with drastic temporal complexity reduction and simulation time.
Anais do XLIX Seminário Integrado de Software e Hardware (SEMISH 2022)
Este artigo contribui para a classificação do tráfego de streaming de vídeo explorando conceitos ... more Este artigo contribui para a classificação do tráfego de streaming de vídeo explorando conceitos de Lógica Fuzzy Intervalar. Essa abordagem estende os trabalhos relacionados ao considerar as incertezas geradas pelas variações nas condições da rede e a imprecisão dos parâmetros que afetam o comportamento do fluxo da rede, o que aumenta a complexidade para alcançar maior acurácia na identificação do tráfego da rede. Algumas avaliações usando a abordagem de lógica intervalar para classificação de tráfego de streaming de vídeo são apresentadas com o uso de aplicações e datasets para validar a proposta.
The main objective of this research is the conception of an approach to predict the risk of morta... more The main objective of this research is the conception of an approach to predict the risk of mortality in ICUs. A cohort of 17,734 patients was used, from the MIMIC-III Database, considering 10 input predictor variables and 8 Machine Learning methods. The best performance was achieved by the Gradient Boosting Machine (GBM) method, which obtained 0,843 (±0,015) of AUC and 0,503 (±0,048) of F1 score. The results are promising and, in some cases, superior to those obtained by other proposals identified in the literature review. Resumo. Esta pesquisa tem por objetivo central a concepção de uma abordagem para predição do risco de mortalidade em UTIs. Foi empregada uma coorte de 17.734 pacientes provenientes do Banco de Dados MIMIC-III, sendo consideradas 10 variáveis preditoras de entrada e 8 métodos de Aprendizado de Máquina. A melhor performance foi alcançada pelo método Gradient Boosting Machine (GBM), que atingiu 0,843 (±0,015) de AUC e 0,503 (±0,048) de F1 score. Os resultados são promissores e, em alguns casos, superiores a outras propostas identificadas na revisão de literatura.
2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Jul 18, 2022
This paper considers the consensual analysis in decision-making (CDM) processes based on fuzzy lo... more This paper considers the consensual analysis in decision-making (CDM) processes based on fuzzy logic (FL) and interval-valued fuzzy logic (IVFL), providing a CDM-strategy, by exploring the axiomatic properties of fuzzy consensus measures (FCM) via penalty functions. Thus, two models are formalized, FS-FCM and IVFS-FCM. In the former, the fuzzy-valued lattice enables the analysis of fuzzy information for linguistic variables (LV), which is obtained by the aggregation of penalty functions. And, in the latter, the consensus measures of fuzzy sets are aggregated to build a new consensual analysis modeling. Thus, e.g., the cohesion of several terms related to the same LV can be analyzed, and also the coherence between fuzzy sets referring to the lowest and highest projections. Such models decide based on relevance criteria and qualitative assessments, via the selection of alternatives, supporting the corresponding algorithmic strategies: FS-FCM strategy, applied to fuzzy values, and IVFS-FCM strategy, covering fuzzy sets. The Intf-HybridMem approach explores the access patterns to volatile and non-volatile memories related to decision-making in two steps: (i) the FS-FCM strategy explores consensus measures of fuzzy values from membership functions; and (ii) the IVFS-FCM strategy, modeling inaccuracy inherent in input variables, as read/write frequency and access recency, also including the migration recommendation as output, which is validated by evaluations carried out in both proposed strategies.
Playful and unplugged activities have been proposed as an improved alternative enabling the inser... more Playful and unplugged activities have been proposed as an improved alternative enabling the insertion of methodological innovation to increase learning in elementary education, through the application of computational concepts and techniques. In such context, the consolidation of the Space Adventure activity is described, presenting the modelling of its final stage named as Task 5. In this stage, the game execution promotes learning via simulation of a scientific space journey. The concepts of abstraction, decomposition, pattern recognition are worked on, and the search for solutions for energy efficiency and correction evaluation of data organization is also encouraged. Resumo. Atividades lúdicas e desplugadas têm sido propostas como alternativas viáveis economicamente e com inserção de inovação metodológica para incremento da aprendizagem na Educação Fundamental, via aplicação de conceitos e técnicas da computação. Neste contexto, descreve-se a consolidação da atividade Aventura Espacial, apresentando a modelagem da sua etapa final, identificada como Tarefa 5. Nesta etapa, a execução das jogadas promove aprendizagem via simulação de uma jornada científica espacial. São trabalhados os conceitos de abstração, decomposição, reconhecimento de padrões e estimulam-se ainda a busca de soluções para eficiência energética e avaliação da correção na organização de dados.
Trends in Applied and Computational Mathematics, Oct 13, 2004
This paper introduces the stochastic version of the Geometric Machine Model for the modelling of ... more This paper introduces the stochastic version of the Geometric Machine Model for the modelling of sequential, alternative, parallel (synchronous) and nondeterministic computations with stochastic numbers stored in a (possibly infinite) shared memory. The programming language L(D → ∞), induced by the Coherence Space of Processes D → ∞ , can be applied to sequential and parallel products in order to provide recursive definitions for such processes, together with a domain-theoretic semantics of the Stochastic Arithmetic. We analyze both the spacial (ordinal) recursion, related to spacial modelling of the stochastic memory, and the temporal (structural) recursion, given by the inclusion relation modelling partial objects in the ordered structure of process construction.
The aim of this paper is to introduce the notion of interval additive generators of interval t-no... more The aim of this paper is to introduce the notion of interval additive generators of interval t-norms as interval representations of additive generators of t-norms, considering both the correctness and the optimality criteria, in order to provide a more systematic methodology for the selection of interval t-norms in the various applications. We prove that interval additive generators satisfy the main
2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Jul 18, 2022
Firstly, this work discusses the main conditions guarantying that general overlap (grouping) func... more Firstly, this work discusses the main conditions guarantying that general overlap (grouping) functions can be obtained from n-dimensional overlap (grouping) functions. Focusing on QL-implications, which are usually generated by strong negations together with t-norms and t-conorms, we consider a non-restrictive construction, by relaxing not only the associativity and the corresponding neutral elements (NE) but also the reverse construction of other properties. Thus, the main properties of the QL-implication class are studied, considering a tuple (G,N,O) generated from grouping and overlap functions together with the greatest fuzzy negation. In addition, in order to provide more flexibility, we define a subclass of QL-implications generated from general overlap and general grouping functions. Some examples are introduced, illustrating the constructive methods to generate such operators.
This paper explores consensus measures to support to determining the level of use of physical mac... more This paper explores consensus measures to support to determining the level of use of physical machines in a cloud computing environment. The fuzzy approach considers the uncertainties present in the cloud computing environment and the consensus analysis of the fuzzy sets is based on arithmetic and exponencial means. In the proposal evaluation, a case study aimed at the architecture of the Framework Int-FLBCC, which is under development by the research group was designed.
A main obstacle in simulation of quantum algorithms is the exponential increase in the temporal a... more A main obstacle in simulation of quantum algorithms is the exponential increase in the temporal and spatial complexities, especially in dense quantum transformations such as the Hadamard operator. In this work, new optimizations for the execution of quantum transformations in the Distributed Geometric Machine (D-GM) environment.Instead of executing them in a single step, they are decomposed and only values different from Identity operator are stored. As a benchmark, Hadamard Transformations were simulated up to 28 qubits in a GPU. When compared to our previous implementation, our new approach is 10, 829⇥ faster and allows for the simulation of more qubits. Resumo. Um dos maiores obstáculos para a simulação de algoritmos
Computer systems based on intuitionistic fuzzy logic are capable of generating a reliable output ... more Computer systems based on intuitionistic fuzzy logic are capable of generating a reliable output even when handling inaccurate input data by applying a rule based system, even with rules that are generated with imprecision. The main contribution of this paper is to show that quantum computing can be used to extend the class of intuitionistic fuzzy sets with respect to representing intuitionistic fuzzy Xor operators. This paper describes a multi-dimensional quantum register using aggregations operators such as t-(co)norms based on quantum gates allowing the modeling and interpretation of intuitionistic fuzzy Xor operations.
International Journal of High Performance Computing Applications, Jan 16, 2019
Exponential increase and global access to read/write memory states in quantum computing (QC) simu... more Exponential increase and global access to read/write memory states in quantum computing (QC) simulation limit both the number of qubits and quantum transformations which can be currently simulated. Although QC simulation is parallel by nature, spatial and temporal complexity are major performance hazards, making this a nontrivial application for high performance computing. A new methodology employing reduction and decomposition optimizations has shown relevant results, but its GPU implementation could be further improved. In this work, we develop a new kernel for in situ GPU simulation that better explores its resources without requiring further hardware. Shor's and Grover's algorithms are simulated up to 25 and 21 qubits respectively and compared to our previous version, to LIQUiji 0 s simulator and to ProjectQ framework, showing better results with relative speedups up to 4.38Â, 3357.76Â and 333Â respectively.
This paper presents seven concurrent hash table implementations in Haskell, ranging from low-leve... more This paper presents seven concurrent hash table implementations in Haskell, ranging from low-level synchronization mechanisms to high-level ones such as transactional memories. The hash tables were compared using different initial sizes, load factors, data types and hash functions. We also present a case study on implementing a color palette algorithm using the hash tables. The result of the comparison between the algorithms shows that the implementation using the STM Haskell transactional memory library and fine-grain synchronization provides the best performance and good scalability, followed by the implementation using lock striping and MVars.
Sustainable Computing: Informatics and Systems, Mar 1, 2021
One of the great challenges of today's computer memory architectures has been their energy consum... more One of the great challenges of today's computer memory architectures has been their energy consumption. In this sense, DRAM technology has come close to its scalability limit due to energy issues. Thus, new technologies have emerged intending to replace DRAM as main memory. In this context, non-volatile memories emerge promising to be an alternative for low energy consumption memory systems. Due to the challenges of these new technologies, hybrid approaches have been a trend. When working with hybrid memories, there is a necessity for decisions in which memory modules each data should be stored. This decision must consider the data access profile and the characteristics of memory technology. Thus, this work presents the Intf-HybridMem architecture, a proposal for page migration in hybrid memories using fuzzy systems to support decision making. The fuzzy approach can model the uncertainties of the data access profile and the characteristics of the memory modules. The tests seek to identify the performance limit of the Intf-HybridMem through the use of an oracle mechanism. Results show that there is potential for exploring migration mechanisms with overhead of 0.18% and 0.39% in some scenarios.
One of the main obstacles for the adoption of quantum algorithm simulation is the exponential inc... more One of the main obstacles for the adoption of quantum algorithm simulation is the exponential increase in temporal and spatial complexities, due to the expansion of transformations and read/write memory states by using tensor product in multi-dimension applications. Reduction and decomposition optimizations via the Id-operator provide a smart and appropriate storage and distribution of quantum information. Reductions are achieved by avoiding replication and sparsity inherited from such operators. By using decompositions, applications may be divided into sub-steps to store only distinct values from Id-operators, instead of executing quantum transformations in a single step. Additional optimizations based on mixed partial processes provide control over increase in read/write memory states in quantum transformations, also contributing to increase the scalability regarding hardware-GPUs memory limit. Hadamard and Discret Quantum Fourier Transforms were simulated up to 28 qubits applications over a single GPU with drastic temporal complexity reduction and simulation time.
Anais do XLIX Seminário Integrado de Software e Hardware (SEMISH 2022)
Este artigo contribui para a classificação do tráfego de streaming de vídeo explorando conceitos ... more Este artigo contribui para a classificação do tráfego de streaming de vídeo explorando conceitos de Lógica Fuzzy Intervalar. Essa abordagem estende os trabalhos relacionados ao considerar as incertezas geradas pelas variações nas condições da rede e a imprecisão dos parâmetros que afetam o comportamento do fluxo da rede, o que aumenta a complexidade para alcançar maior acurácia na identificação do tráfego da rede. Algumas avaliações usando a abordagem de lógica intervalar para classificação de tráfego de streaming de vídeo são apresentadas com o uso de aplicações e datasets para validar a proposta.
Uploads
Papers by Renata Reiser