CV Lattes
ResearcherID
ORCID
Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação (FEEC) (Institutional affiliation from the last research proposal) Birthplace: Brazil
João Marcos Travassos Romano was born in Rio de Janeiro on April 12, 1960. He graduated in Electrical Engineering at University of Campinas (1981), he got his M.Sc. in Electrical Engineering from the University of Campinas (1984) and his Ph.D. in Signal Processing and Automation from the Université de Paris XI, Orsay (1987). He is currently Full Professor at the School of Electrical and Computer Engineering at the University of Campinas (FEEC / UNICAMP), where he has been a full-time professor since 1988. Since April 2021, he is the Vice-Dean for Research at UNICAMP. He was Associate Director (2003-2007) and Director (2015-2019) of FEEC / UNICAMP. He was the President of the Brazilian Telecommunications Society (SBrT) from 2000 to 2004. He received from UNICAMP the Academic Recognition Award "Zeferino Vaz", 2008 edition, for highlighting teaching and research activities, as well as the PROEC Award, for highlighting extension activities, in the 2020 edition. He was a Visiting Professor at the Conservatoire National des Arts et Metier (CNAM / Paris), at the Université de Paris-Descartes and at the École Normale Supérieure de Cachan, France. He is co-author of Unsupervised Signal Processing: Channel Equalization and Source Separation, edited by CRC Press (Taylor and Francis Group LLC) in 2011. He is an Emeritus Member of SBrT and a Senior Member of the IEEE. His main research interests are related to the theory and methods of signal processing and filtering, machine learning, and several potential applications and related topics, such as telecommunications, seismic, blind source separation, recommendation systems and decision support etc. Since April 2021, he has been the Director of the BI0S-Brazilian Institute of Data Science, a Center for Applied Research in Artificial Intelligence approved under the Scientific and Technological Cooperation Agreement between FAPESP and MCTIC / CGI. (Source: Lattes Curriculum)
| Articles published in Pesquisa FAPESP Magazine about the researcher: |
| La inteligencia artificial llega a la salud en Brasil |
| Artificial intelligence in healthcare |
| Articles published in Pesquisa para Inovação FAPESP about the researcher: |
| Six new artificial intelligence research centers announced in call issued by FAPESP with partners |
| News published in Agência FAPESP Newsletter about the researcher |
| TITULO |
| Articles published in other media outlets (0 total): |
| More itemsLess items |
| VEICULO: TITULO (DATA) |
| VEICULO: TITULO (DATA) |
The Brazilian Institute of Data Science (BI0S) is a research center with the main purpose to develop state-of-the-art solutions in data science and artificial intelligence (AI), solving relevant problems and connecting academia, companies, startups, society and the public sector in an integrated innovation ecosystem. It will contribute to scientific, technological and social development, …
This project makes use of the institutional technical reserves and aims to improve channel the available resources of the research infrastructure of FEEC. It aims to maintain and improve the collective infrastructure support research and it is focused on the computer network. The FEEC community decided to use the FAPESP RTI resources plan in the wireless network reform of the building blo…
Prof. Daniela Donno is world-leading expert in the use of advanced signal processing techniques for processing geophysical and seismic signals, an area in which our research group has been greatly interested in the past years. The goal of geophysical signal processing is to estimate the properties of the Earth's subsurface, based on measurements acquired by sensors located over the area t…
(Only some records are available in English at this moment)
We define a signal, in a broad sense, as an entity that carries information. To better analyze the information of interest, a range of techniques have been developed, especially the ones based on signal decomposition. These techniques are used according to the features we want to extract. Formally, they are implemented by means of matricial or tensorial decomposition. In our MSc research …
One of the most traditional problems in signal processing is the deconvolution, in which the objective is to retrieve a signal that has been distorted by a given system. This problem is still very challenging in its unsupervised version, where the retrieval is carried without knowledge of the system and with access only to some information about the original signal's structure. Among many…
Decision making problems involving multiple criteria can be found in several practical situations. Many of these problems have specific characteristics that can lead to biased results, especially with respect to the size of the problem data. For example, when there is a large number of criteria, many of them may not be necessary in terms of expressiveness in the model, or even be correlat…
(Only some records are available in English at this moment)
Machine Learning Models have received a lot of attention in a variety of problems due to their great performances and ability to learn from data. But the application of such models in social-economic problems---credit assessment, recidivism prediction, students selection, among others--- has raised ethical concerns. Biased data usually leads the model to biased decisions, which will produ…
The present project is composed of two lines of research that have in common an inverse problem formulation, to be solved by means of signal processing methods, particularly those based on sparse estimation.In the first line of research, we intend to continue our work on blind source separation, exploring more deeply the relationships between the criteria employed and the sparsity and ant…
Several practical situations can be modeled as multi-criteria decision making (MCDM) problems. Generally, methods used to deal with MCDM problems do not take into account the data structural information, such as the interaction between criteria. For example, correlated criteria may bias the decision analysis towards a specific alternative. In this context, an approach based on Choquet int…
Deep learning neural networks have attracted the attention of the machine learning community because of their appealing data-driven framework and of their performance in a number of pattern recognition tasks. It is well known that the state-of-the-art Deep Neural Networks (DNNs) are highly redundant and contain hundreds of millions of parameters, using up all available memory of personal …
| 1 / 1 | Ongoing research grants |
| 12 / 9 | Completed research grants |
| 28 / 9 | Completed scholarships in Brazil |
| 1 / 1 | Ongoing scholarships abroad |
| 3 / 3 | Completed scholarships abroad |
| 45 / 23 | All research grants and scholarships |
| Associated processes | |