Articles — Two-column

El análisis estadístico de alta dimensión y tamaño de muestra pequeño (HDLSS) se está aplicando cada vez más en una amplia gama de contextos. En tales situaciones, se ve que el popular método de la Máquina de Vectores Soporte (SVM) sufre de ''Acumulación de datos'' en el margen, lo que puede disminuir la capacidad de generalización del modelo. Esto conduce al desarrollo de la Distance Weighted Discrimination para encontrar un hiperplano separador . En el presente trabajo se revisa y reproduce, con detalle en la derivación y solución de la función de pérdida que se resuelve usando SOCP, del método desarrollado en e implementado en el entorno R\cite{R}. Basado en el trabajo e implementación de se aplica y comparan resultados a conjuntos de datos reales y simulados (en medida de lo posible los mismos conjuntos de datos utilizados que en) Palabras clave: SVM, kernel, R (el ambiente de cómputo estadístico) y datos de alta dimensión con tamaño de muestra pequeño (data High Dimension Low Sample Size).

Lazaridou., et al 2017 proposed a framework for language learning that relies on multi-agent communication. The agents in the framework were setup in a referential game where they communicated about many images. In this paper, we propose an experiment where agents develop a private language for referring to specified sentences given a set of sentences. The challenge is for the agents to learn a method of distinguishing differences between sentences and to develop a shared language to be able to refer to particular sentences by those distinguishing features. We will evaluate the agents' ability to accurately identify and differentiate the sentences. In addition, we will identify patterns in the methods that the agents develop to refer to the different types of sentences.Keywords: Reinforcement learning, multi-agent coordination

Um artigo sobre controle digital de conversores CC-CC tipo buck utilizando microcontrolador STM cortex-m0+.

In this paper, we evaluate a baseline word embedding model for a set of clinical notes derived from patient records. For our baseline, we extract features for this embedding using the Word2Vec module from the gensim package. We also build two models, a word2vec skipgram model with negative sampling and a positive point-wise mutual information (PPMI) model by training on the processed clinical notes. Our evaluation shows that both the PPMI and the skipgram models show improved results for medically-related terms when compared with the baseline model. PPMI shows the best result out of all three models.