Repositório Comunidade: CCTCCCTChttps://hdl.handle.net/1822/212902024-03-28T19:44:11Z2024-03-28T19:44:11ZA FIPA-compliant framework for integrating rule engines into software agents for supporting communication and collaboration in a multiagent platformAguayo-Canela, Francisco J.Alaiz-Moretón, HéctorGarcía-Rodríguez, IsaíasBenavides-Cuéllar, CarmenBenítez-Andrades, José AlbertoNovais, Paulohttps://hdl.handle.net/1822/863792023-09-14T19:53:19ZTítulo: A FIPA-compliant framework for integrating rule engines into software agents for supporting communication and collaboration in a multiagent platform
Autor: Aguayo-Canela, Francisco J.; Alaiz-Moretón, Héctor; García-Rodríguez, Isaías; Benavides-Cuéllar, Carmen; Benítez-Andrades, José Alberto; Novais, Paulo
Resumo: Production rules have been traditionally considered a good knowledge representation formalism for creating expert systems, and also as a good mechanism for building intelligence within software agents due to the flexibility of their declarative knowledge representation. But the use of rules and rule engines for implementing behaviors inside agents in the context of a multiagent environment has a number of challenges if one wants to adhere and preserve the defining characteristics of agency and support the kind of communication and collaboration mechanisms needed by multiagent systems. This paper describes a framework, and its implementation, for the integration of production rules reasoning mechanisms inside software agents in the context of a multiagent platform, preserving the defining characteristics of agency and accomplishing the well-established FIPA standards for agent interaction.
The implementation has been accomplished by using either CLIPS or Jess as the rule engine, but it can be extended to other formalisms. JADE was chosen as the multiagent platform for developing and testing the solution. The proposed framework includes the uncoupled integration of the rule inference engine into the agent and the adherence to FIPA specifications about using protocols and communication processes. The resulting framework can be used to adapt legacy expert systems whose functionality can be divided to be performed by separate agents or to build new, distributed intelligent systems exploiting the capacities of a multiagent platform at different levels.
To show the validity of the approach, a functional test environment was built and is available for downloading.
<b>Tipo</b>: conferencePaperARCANO - No caminho para adicionar semântica às descrições arquivísticasCunha, Luís FilipeRamalho, José Carloshttps://hdl.handle.net/1822/860732023-08-09T20:48:47Z2023-08-09T20:48:47ZTítulo: ARCANO - No caminho para adicionar semântica às descrições arquivísticas
Autor: Cunha, Luís Filipe; Ramalho, José Carlos
Resumo: Nos últimos anos, o uso de técnicas de Machine Learning no Processamento de Linguagem Natural tem sido recorrente no processamento de grandes quantidades de documentos não estruturados. No entanto, nem sempre é fácil encontrar recursos de língua portuguesa, nomeadamente datasets anotados, que permitam fazer uso destas tecnologias. Neste trabalho é apresentada uma ferramenta de anotação inteligente ARCANO, que faz uso de algoritmos de Machine Learning para assistir o processo de anotação de documentos portugueses. Esta ferramenta foi especializada e treinada no domínio das descrições arquivísticas.
Descrição: http://ner.epl.di.uminho.pt/
<b>Tipo</b>: conferencePaper2023-08-09T20:48:47ZOptimized voronoi-based algorithms for parallel shortest vector computationMariano, ArturCabeleira, FilipeSantos, Luís PauloFalcão, Gabrielhttps://hdl.handle.net/1822/781352022-05-31T15:37:22Z2022-05-31T15:37:22ZTítulo: Optimized voronoi-based algorithms for parallel shortest vector computation
Autor: Mariano, Artur; Cabeleira, Filipe; Santos, Luís Paulo; Falcão, Gabriel
Resumo: This chapter addresses Voronoi cell-based algorithms, solving the Shortest Vector Problem, a fundamental challenge in lattice-based cryptanalysis. Several optimizations reduce the original algorithm's execution time. The algorithm suitability for parallel execution on both CPUs and GPUs is also shown. Optimizations are based on pruning, avoiding computations that will not improve the solution. The pruning criteria is related to the target vectors norm relative to the current best solution vector norm. Speedups up to 69× are observed. With a pre-process sorting step, which requires storing the norm ordered target vectors and therefore significantly more memory, speedup increases to 77×. On the parallel processing side, the optimized algorithm exhibits linear scalability on a CPU with up to 28 threads and keeps scaling, at a lower rate, with Simultaneous Multi-Threading up to 56 threads. The lack of support for efficient threads synchronization in GPUs precludes a scalable implementation of the pruning optimization. A parallel GPU version of the non-optimized algorithm is demonstrated to be competitive with the parallel non optimized CPU version, although the latter outperforms the former for 56 threads. GPUs are expected to outperform CPUs for higher lattice dimensions; this cannot be experimentally verified due to the limited memory available on current GPUs.
<b>Tipo</b>: bookPart2022-05-31T15:37:22ZForeword to the special section on recent advances in graphics and interactionRodrigues, NunoMendes, DanielSantos, Luís PauloBouatouch, Kadihttps://hdl.handle.net/1822/781322024-03-27T15:51:16Z2022-05-31T14:52:42ZTítulo: Foreword to the special section on recent advances in graphics and interaction
Autor: Rodrigues, Nuno; Mendes, Daniel; Santos, Luís Paulo; Bouatouch, Kadi
Resumo: This special section on Recent Advances in Graphics and In-
teraction features the papers submitted to Computers & Graphics
(C&G) and presented at the 2021 edition of the International
Conference on Graphics and Interaction – ICGI’2021 – which was
held on November 4 and 5, 2021 at the Faculty of Engineering of
the University of Porto, Portugal, as a joint organization with the
Eurographics Portuguese Chapter — GPCG.
<b>Tipo</b>: journalEditorial2022-05-31T14:52:42ZEnsemble metropolis light transportBashford-Rogers, ThomasSantos, Luís PauloMarnerides, DemetrisDebattista, Kurthttps://hdl.handle.net/1822/781312022-05-31T16:34:59Z2022-05-31T14:28:49ZTítulo: Ensemble metropolis light transport
Autor: Bashford-Rogers, Thomas; Santos, Luís Paulo; Marnerides, Demetris; Debattista, Kurt
Resumo: This article proposes a Markov Chain Monte Carlo (MCMC) rendering algorithm based on a family of guided transition kernels. The kernels exploit properties of ensembles of light transport paths, which are distributed according to the lighting in the scene, and utilize this information to make informed decisions for guiding local path sampling. Critically, our approach does not require caching distributions in world space, saving time and memory, yet it is able to make guided sampling decisions based on whole paths. We show how this can be implemented efficiently by organizing the paths in each ensemble and designing transition kernels for MCMC rendering based on a carefully chosen subset of paths from the ensemble. This algorithm is easy to parallelize and leads to improvements in variance when rendering a variety of scenes.
<b>Tipo</b>: article2022-05-31T14:28:49Z