Madeira Group of Knowledge Representation and Reasoning
Projects
[2018 – 2021] BRANT: Belief Revision applied to Neurorehabilitation Therapy
PI: Eduardo Fermé
Sponsors: Fundação para a Ciência e a Tecnología (FCT). Ministério da Educação e Ciência de Portugal.
PTDC/CCI-COM/30990/2017
Cognitive deficits are common after brain injury, dementia and normal cognitive decline due to aging. These impact the performance of activities of daily living and limit people’s independence, with a high monetary and societal cost. Cognitive rehabilitation has been shown to be the most effective way to address this problem. However, current rehabilitation has some limitations: a) Rehabilitation tools are not adaptive, and may not be adequate for every patient. A suboptimal set of exercises limits the impact of rehabilitation and reduces engagement in rehabilitation. b) Interventions are time consuming and have a high cost, and are usually implemented in clinical environments. BRaNT is an interdisciplinary effort to create a new set of ICT for at home rehabilitation that addresses scientific limitations of current practices and provides solutions for the sustainability of health systems and contributes to the improvement of quality of life of patients.
[2019 – 2020] Advanced belief change operations based on comparisons and conditionals: Towards a general framework
Sponsors: German-Portuguese project. DAAD – Germany- Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
German Co-Chair: Hans Rott
This project aims at developing a general framework of multiple belief change that is able to also take the credibility of information into regard, starting with total preorders as a commonly accepted base of change theory, and combining two methodologies of belief change that build upon total preorders to express knowledge: revision by comparison and c-change theory.
[2010-2011] Implementation of Belief Revision Algorithms
Sponsors: Argentinean-Portuguese project. Secretaría de Ciencia, Tecnología, e Innovación Productiva. Gobierno de Argentina. Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
Argentinean Co-Chair: Ricardo Rodríguez.
Abstract: We propose to develop algorithms aimed to perform belief change in resource bounded agents. In order to do that, we must first try to answer the following research questions: 1. “Which of the constructive models is more suitable for implementations?” 2. “How can the models proposed in the 1980’s be evaluated at the light of the AGM theory? Which properties can we rescue/reuse?” 3. How can we move (back) from the knowledge level to the symbolic level? How can we define the metrics to evaluate the different algorithms?
[2010] Improvement belief change operators with limits of credibility
Sponsors: French-Portuguese project. CNRS – France- Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
French Co-Chair: Sébastien Konieczny
Abstract: The purpose of the project is defining a new model of belief revision called “improvement operator with a limit of credibility”. This operator will be work in the following way: Every time than an rational agent receives a new information (typically a formula) this formula will gain in plausibility in the epistemic state of the agent. And if the agent receives the same new information many times, such that reach the status of a credible sentence, then he will finally believe it. This operator will combine two previous models proposed in the literature: Credibility-Limited functions, proposed by (Hansson, Fermé, Cantwell and Falappa 2001), by (Fermé and Hansson 2001) and by (Fermé, Mikalef and Taboada 2003) and Improvement Operators, proposed by (Konieczny and Pino Perez 2009).
[2008/2009] Knowledge Dynamics in Cognitive Agents Sponsors: Argentinean-Portuguese project.
Sponsors: Secretaría de Ciencia, Tecnología, e Innovación Productiva. Gobierno de Argentina. Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
Argentinean Co-Chair: Marcelo Falappa. Abstract: The project proposes a Logical axiomatisation for belief bases functions (like Levi contraction, Semi-Contraction), how to extend the results for belief bases of partial meet contraction and kernel contraction for supplementary postulates and how to elucidate the relation among different kinds of supplementary postulates for belief bases.
[2018 – 2021] BRANT: Belief Revision applied to Neurorehabilitation Therapy
PI: Eduardo Fermé
Sponsors: Fundação para a Ciência e a Tecnología (FCT). Ministério da Educação e Ciência de Portugal.
PTDC/CCI-COM/30990/2017
Cognitive deficits are common after brain injury, dementia and normal cognitive decline due to aging. These impact the performance of activities of daily living and limit people’s independence, with a high monetary and societal cost. Cognitive rehabilitation has been shown to be the most effective way to address this problem. However, current rehabilitation has some limitations: a) Rehabilitation tools are not adaptive, and may not be adequate for every patient. A suboptimal set of exercises limits the impact of rehabilitation and reduces engagement in rehabilitation. b) Interventions are time consuming and have a high cost, and are usually implemented in clinical environments. BRaNT is an interdisciplinary effort to create a new set of ICT for at home rehabilitation that addresses scientific limitations of current practices and provides solutions for the sustainability of health systems and contributes to the improvement of quality of life of patients.
[2019 – 2020] Advanced belief change operations based on comparisons and conditionals: Towards a general framework
Sponsors: German-Portuguese project. DAAD – Germany- Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
German Co-Chair: Hans Rott
This project aims at developing a general framework of multiple belief change that is able to also take the credibility of information into regard, starting with total preorders as a commonly accepted base of change theory, and combining two methodologies of belief change that build upon total preorders to express knowledge: revision by comparison and c-change theory.
[2010-2011] Implementation of Belief Revision Algorithms
Sponsors: Argentinean-Portuguese project. Secretaría de Ciencia, Tecnología, e Innovación Productiva. Gobierno de Argentina. Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
Argentinean Co-Chair: Ricardo Rodríguez.
Abstract: We propose to develop algorithms aimed to perform belief change in resource bounded agents. In order to do that, we must first try to answer the following research questions: 1. “Which of the constructive models is more suitable for implementations?” 2. “How can the models proposed in the 1980’s be evaluated at the light of the AGM theory? Which properties can we rescue/reuse?” 3. How can we move (back) from the knowledge level to the symbolic level? How can we define the metrics to evaluate the different algorithms?
[2010] Improvement belief change operators with limits of credibility
Sponsors: French-Portuguese project. CNRS – France- Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
French Co-Chair: Sébastien Konieczny
Abstract: The purpose of the project is defining a new model of belief revision called “improvement operator with a limit of credibility”. This operator will be work in the following way: Every time than an rational agent receives a new information (typically a formula) this formula will gain in plausibility in the epistemic state of the agent. And if the agent receives the same new information many times, such that reach the status of a credible sentence, then he will finally believe it. This operator will combine two previous models proposed in the literature: Credibility-Limited functions, proposed by (Hansson, Fermé, Cantwell and Falappa 2001), by (Fermé and Hansson 2001) and by (Fermé, Mikalef and Taboada 2003) and Improvement Operators, proposed by (Konieczny and Pino Perez 2009).
[2008/2009] Knowledge Dynamics in Cognitive Agents Sponsors: Argentinean-Portuguese project.
Sponsors: Secretaría de Ciencia, Tecnología, e Innovación Productiva. Gobierno de Argentina. Ministério da Ciência e do Ensino Superior. Governo de Portugal.
Portuguese Co-Chair: Eduardo Fermé
Argentinean Co-Chair: Marcelo Falappa. Abstract: The project proposes a Logical axiomatisation for belief bases functions (like Levi contraction, Semi-Contraction), how to extend the results for belief bases of partial meet contraction and kernel contraction for supplementary postulates and how to elucidate the relation among different kinds of supplementary postulates for belief bases.