Configuring databases for efficient querying is a complex task, often carried out by a database administrator. Solving the problem of building indexes that truly optimize database access requires a ...substantial amount of database and domain knowledge, the lack of which often results in wasted space and memory for irrelevant indexes, possibly jeopardizing database performance for querying and certainly degrading performance for updating. In this paper, we develop the
SmartIX
architecture to solve the problem of automatically indexing a database by using reinforcement learning to optimize queries by indexing data throughout the lifetime of a database. We train and evaluate
SmartIX
performance using TPC-H, a standard, and scalable database benchmark. Our empirical evaluation shows that
SmartIX
converges to indexing configurations with superior performance compared to standard baselines we define and other reinforcement learning methods used in related work.
Smart Makerspace: A Web Platform Implementation Licks, Gabriel; Teixeira, Adriano; Luyten, Kris
International journal of emerging technologies in learning,
01/2018, Letnik:
13, Številka:
2
Journal Article
Recenzirano
Odprti dostop
Makerspaces are creative and learning environments, home to activities such as fabrication processes and Do-It-Yourself (DIY) tasks. However, containing equipment that are not commonly seen or ...handled, these spaces can look rather challenging to novice users. This paper is based on the Smart Makerspace research from Autodesk, which uses a smart workbench for an immersive instructional space for DIY tasks. Having its functionalities in mind and trying to overcome some of its limitations, we approach the concept building an immersive instructional space as a web platform. The platform, introduced to users in a makerspace, had a feedback that reflects its potential between novice and intermediate users, for creating facilitators and encouraging these users.
Goal recognition is the task of identifying the goal an observed agent is pursuing. The quality of its results depends on the quality of the observed information. In most goal recognition approaches, ...the accuracy significantly decreases in settings with missing observations. To mitigate this issue, we develop a learning model based on LSTMs, leveraging attention mechanisms, to enhance observed traces by predicting missing observations in goal recognition problems. We experiment using a dataset of goal recognition problems and apply the model to enhance the observation traces where missing. We evaluate the technique using a state-of-the-art goal recognizer in four different domains to compare the accuracy between the standard and the enhanced observation traces. Experimental evaluation shows that recurrent neural networks with self-attention mechanisms improve the accuracy metrics of state-of-the-art goal recognition techniques by an average of 60%.
RESUMO Este artigo apresenta os resultados de uma pesquisa realizada no projeto Escola de Hackers cujo objetivo é oportunizar o aprendizado de técnicas e habilidades de programação para alunos do ...ensino fundamental. Mais especificamente, propõe-se a analisar a forma como a programação de computadores provoca processos de recuperação e consciência, mecanismos auxiliares da aprendizagem segundo o aporte teórico de Juan Ignacio Pozo (2002). Para tanto procedeu-se a uma pesquisa de cunho qualitativo e exploratório, com triangulação na coleta de dados primários entre observação, aplicação de questionários e entrevistas, junto a uma amostra composta por 11 alunos de uma escola municipal de Ensino Fundamental de Passo Fundo/RS, participantes do projeto no ano de 2016. Os principais resultados obtidos permitem inferir que os alunos utilizaram os processos de recuperação e consciência para programar. Com ênfase, evidenciou-se maior incidência de uso das subcategorias reconhecimento e transferência, apontadas como os principais recursos metacognitivos utilizados pelos estudantes em suas tarefas. Palavras-chave: Processos auxiliares da aprendizagem. Recuperação e consciência. Ensino fundamental. Metacognição. Abstract This paper presents the results of a research carried out in the School of Hackers project, whose objective is to provide elementary school students with the opportunity to learn programming techniques and skills. More specifically, it proposes to analyze how computer programming provokes the processes of recovery and awareness, which are auxiliary mechanisms of learning, according to the theoretical contribution of Juan Ignacio Pozo (2002). For that, a qualitative and exploratory research was carried out, with triangulation in the primary data collection between observation, questionnaires application and interviews, along with a sample composed by 11 elementary school students from an elementary school of Passo Fundo/RS, who participated in the project in the year of 2016. The main results obtained allow us to infer that the students used, while programming, the recovery and awareness processes. With emphasis, it was evidenced a higher incidence of the use of recognition and transfer subcategories, pointed out as the main metacognitive resources used by the students in their tasks. Keywords: Auxiliary processes of learning. Recovery and awareness. Elementary school. Metacognition. RESUMEN Este artículo presenta los resultados de una encuesta realizada en el proyecto Escuela de Hackers cuyo objetivo es oportunizar el aprendizaje de técnicas y habilidades de programación para alumnos de la enseñanza fundamental. Más específicamente, se propone analizar la forma en que la programación de computadoras provoca procesos de recuperación y conciencia, mecanismos auxiliares del aprendizaje según el aporte teórico de Juan Ignacio Pozo (2002). Para ello se procedió a una investigación de cuño cualitativo y exploratorio, con triangulación en la recolección de datos primarios entre observación, aplicación de cuestionarios y entrevistas, junto a una muestra compuesta por 11 alumnos de una escuela municipal de Enseñanza Fundamental de Passo Fundo / RS , participantes del proyecto en el año 2016. Los principales resultados obtenidos permiten inferir que los alumnos utilizaron los procesos de recuperación y conciencia para programar. Con énfasis, se evidenció mayor incidencia de uso de las subcategorías reconocimiento y transferencia, apuntadas como los principales recursos metacognitivos utilizados por los estudiantes en sus tareas. Palabras clave: Procesos auxiliares del aprendizaje. Recuperación y consciência. Enseñanza fundamental. Metacognición. DOI: http://dx.doi.org/10.22169/revint.v13i30.1497
Configuring databases for efficient querying is a complex task, often carried out by a database administrator. Solving the problem of building indexes that truly optimize database access requires a ...substantial amount of database and domain knowledge, the lack of which often results in wasted space and memory for irrelevant indexes, possibly jeopardizing database performance for querying and certainly degrading performance for updating. We develop an architecture to solve the problem of automatically indexing a database by using reinforcement learning to optimize queries by indexing data throughout the lifetime of a database. In our experimental evaluation, our architecture shows superior performance compared to related work on reinforcement learning and genetic algorithms, maintaining near-optimal index configurations and efficiently scaling to large databases.
Our work aims at developing reinforcement learning algorithms that do not rely on the Markov assumption. We consider the class of Non-Markov Decision Processes where histories can be abstracted into ...a finite set of states while preserving the dynamics. We call it a Markov abstraction since it induces a Markov Decision Process over a set of states that encode the non-Markov dynamics. This phenomenon underlies the recently introduced Regular Decision Processes (as well as POMDPs where only a finite number of belief states is reachable). In all such kinds of decision process, an agent that uses a Markov abstraction can rely on the Markov property to achieve optimal behaviour. We show that Markov abstractions can be learned during reinforcement learning. Our approach combines automata learning and classic reinforcement learning. For these two tasks, standard algorithms can be employed. We show that our approach has PAC guarantees when the employed algorithms have PAC guarantees, and we also provide an experimental evaluation.
A pré-eclâmpsia é definida como uma disfunção multissistêmica proveniente de uma malformação das artérias espirais provocando hipertensão, proteinúria e edema. Sua explicação fisiopatológica é que ...essa malformação provoca estresse oxidativo na placenta que consequentemente secretará fatores pró-inflamatórios vasculares, causando disfunção endotelial suscitando dessa forma, os sintomas da doença. Se o diagnóstico e tratamento não forem precisos pode haver complicações, sendo a eclâmpsia a principal delas, que é caracterizada como uma emergência obstétrica com convulsões tônico-clônicas. O tratamento para essa enfermidade é feito com medicamentos sintomáticos, uma vez que o tratamento definitivo é o parto. Portanto, o seguinte artigo visa elucidar os mecanismos fisiopatológicos, suas implicações clínicas e os demais temas oriundos desta patologia.