[pt] Spoken Dialogue Systems (SDS s) são sistemas baseados em computadores desenvolvidos para fornecerem informações e realizar tarefas utilizando o diálogo como forma de interação. Eles são capazes de reconhecimento de voz, interpretação, gerenciamento de diálogo e são capazes de ter uma voz como saída de dados, tentando reproduzir uma interação natural falada entre um usuário humano e um sistema. SDS s provém diferentes serviços, todos através de linguagem falada com um sistema. Mesmo com todo o desenvolvimento nesta área, há escassez de informações sobre como avaliar a qualidade de tais sistemas com o propósito de otimização do mesmo. Com dois destes sistemas, BoRIS e INSPIRE, usados para reservas de restaurantes e gerenciamento de casas inteligentes, diversos experimentos foram conduzidos no passado, onde tais sistemas foram utilizados para resolver tarefas específicas. Os participantes avaliaram a qualidade do sistema em uma série de questões. Além disso, todas as interações foram gravadas e anotadas por um especialista.O desenvolvimento de métodos para avaliação de performance é um tópico aberto de pesquisa na área de SDS s. Seguindo a idéia do modelo PARADISE (PARAdigm for DIalogue System Evaluation – desenvolvido pro Walker e colaboradores na AT&T em 1998), diversos experimentos foram conduzidos para desenvolver modelos de previsão de performance de sistemas de reconhecimento de voz e linguagem falada. O objetivo desta dissertação de mestrado é desenvolver modelos que permitam a previsão de dimensões de qualidade percebidas por um usuário humano, baseado em parâmetros instrumentalmente mensuráveis utilizando dados coletados nos experimentos realizados com os sistemas BoRIS e INSPIRE , dois sistemas de reconhecimento de voz (o primeiro para busca de restaurantes e o segundo para Smart Homes). Diferentes algoritmos serão utilizados para análise (Regressão linear, Árvores de Regressão, Árvores de Classificação e Redes Neurais) e para cada um dos algoritmos, uma ferramenta diferente será programada em MATLAB, para poder servir de base para análise de experimentos futuros, sendo facilmente modificado para sistemas e parâmetros novos em estudos subsequentes.A idéia principal é desenvolver ferramentas que possam ajudar na otimização de um SDS sem o envolvimento direto de um usuário humano ou servir de ferramenta para estudos futuros na área. / [en] Spoken Dialogue Systems (SDS s) are computer-based systems developed to provide information and carry out tasks using speech as the interaction mode. They are capable of speech recognition, interpretation, management of dialogue and have speech output capabilities, trying to reproduce a more or less natural spoken interaction between a human user and the system. SDS s provide several different services, all through spoken language. Even with all this development, there is scarcity of information on ways to assess and evaluate the quality of such systems with the purpose of optimization. With two of these SDS s ,BoRIS and INSPIRE, (used for Restaurant Booking Services and Smart Home Systems), extensive experiments were conducted in the past, where the systems were used to resolve specific tasks. The evaluators rated the quality of the system on a multitude of scales. In addition to that, the interactions were recorded and annotated by an expert. The development of methods for performance evaluation is an open research issue in this area of SDS s. Following the idea of the PARADISE model (PARAdigm for DIalogue System Evaluation model, the most well-known model for this purpose (developed by Walker and co-workers at AT&T in 1998), several experiments were conducted to develop predictive models of spoken dialogue performance. The objective of this dissertation is to develop and assess models which allow the prediction of quality dimensions as perceived by the human user, based on instrumentally measurable variables using all the collected data from the BoRIS and INSPIRE systems. Different types of algorithms will be compared to their prediction performance and to how generic they are. Four different approaches will be used for these analyses: Linear regression, Regression Trees, Classification Trees and Neural Networks. For each of these methods, a different tool will be programmed using MATLAB, that can carry out all experiments from this work and be easily modified for new experiments with data from new systems or new variables on future studies. All the used MATLAB programs will be made available on the attached CD with an operation manual for future users as well as a guide to modify the existing programs to work on new data. The main idea is to develop tools that would help on the optimization of a spoken dialogue system without a direct involvement of the human user or serve as tools for future studies in this area.

<p>Convolutional neural networks (CNNs) have become important workloads due to their<br> impressive accuracy in tasks like image classification and recognition. Convolution operations<br> are compute intensive, and this cost profoundly increases with newer and better CNN models.<br> However, convolutions come with characteristics such as sparsity which can be exploited. In<br> this dissertation, we propose three different works to capture sparsity for faster performance<br> and reduced energy. </p> <p><br></p> <p>The first work is an accelerator design called <em>SparTen</em> for improving two-<br> sided sparsity (i.e, sparsity in both filters and feature maps) convolutions with fine-grained<br> sparsity. <em>SparTen</em> identifies efficient inner join as the key primitive for hardware acceleration<br> of sparse convolution. In addition, <em>SparTen</em> proposes load balancing schemes for higher<br> compute unit utilization. <em>SparTen</em> performs 4.7x, 1.8x and 3x better than dense architecture,<br> one-sided architecture and SCNN, the previous state of the art accelerator. The second work<br> <em>BARISTA</em> scales up SparTen (and SparTen like proposals) to large-scale implementation<br> with as many compute units as recent dense accelerators (e.g., Googles Tensor processing<br> unit) to achieve full speedups afforded by sparsity. However at such large scales, buffering,<br> on-chip bandwidth, and compute utilization are highly intertwined where optimizing for<br> one factor strains another and may invalidate some optimizations proposed in small-scale<br> implementations. <em>BARISTA</em> proposes novel techniques to balance the three factors in large-<br> scale accelerators. <em>BARISTA</em> performs 5.4x, 2.2x, 1.7x and 2.5x better than dense, one-<br> sided, naively scaled two-sided and an iso-area two-sided architecture, respectively. The last<br> work, <em>EUREKA</em> builds an efficient tensor core to execute dense, structured and unstructured<br> sparsity with losing efficiency. <em>EUREKA</em> achieves this by proposing novel techniques to<br> improve compute utilization by slightly tweaking operand stationarity. <em>EUREKA</em> achieves a<br> speedup of 5x, 2.5x, along with 3.2x and 1.7x energy reductions over Dense and structured<br> sparse execution respectively. <em>EUREKA</em> only incurs area and power overheads of 6% and<br> 11.5%, respectively, over Ampere</p>

<p dir="ltr">Powder Sprays such as Direct Energy Deposition and Cold Spray are rapidly growing and promising manufacturing methods in the Additive Manufacturing field, as they allow easy and localized delivery of powder to be fused to a substrate and consecutive layers. The relatively small size of nozzles allows for these methods to be mounted on CNC machines and Robotic Arms for the creation of complex shapes. However, these manufacturing methods are inherently stochastic, and therefore differences in powder size, shape, trajectory, and velocity can drastically affect whether they will deposit on a substrate. This variation results in an inherent reduction of deposition efficiency, leading to waste and the need for powder collection or recycling systems. The design of the nozzles can drastically affect the variation of powder trajectory and velocity on a holistic level, and thus understanding the gas-powder flow of these nozzles in respect to the features of said nozzles is crucial. This paper proposes and examines how changes in the nozzle geometry affect gas-powder flow and powder focusing for Direct Energy Deposition and Cold Spray. In addition, a new Pulsed Cold Spray nozzle design is proposed that will control the amount of gas and powder used by the nozzle via solenoid actuation. By making these changes to the nozzle, it is possible to improve deposition efficiency and reduce powder/gas waste in these processes, while also allowing for improved coating density. Furthermore, the research done in this thesis will also focus on novel applications to powder spray manufacturing methods, focusing on polymer metallization and part identification.</p>

<p dir="ltr">Ultra-scaled, always-on, smart, wearable and implantable (WI) therapeutic platforms define the research frontier of modern personalized medicine. The WI platform integrates real-time sensing with on-demand therapy and is ideally suited for real-time management of chronic diseases like diabetes. Traditional blood tracking methods, such as glucometers, are insufficient due to their once-in-a-while measurements and the imprecision of insulin injections, which can lead to severe complications. To address these challenges, researchers have been developing smart and minimally invasive microneedle (MN) components for pain-free glucose detection and drug delivery, potentially functioning as an "artificial pancreas". Inspired by natural body homeostasis, these platforms must be accurate and responsive for immediate corrective interventions. However, artificial MN patches often have slow readings due to factors like MN morphology and composition that remain poorly understood, hindering their optimization and integration into real-time monitoring devices. Despite extensive, iterative experimental efforts worldwide, a holistic framework incorporating the interaction between MN sensing and therapy with fluctuating natural body functions is missing. In this thesis, we propose a generalized framework for glycemic management based on the interaction between biological processes and MN-based operations. The results, incorporating theoretical insights from the 1960s and recent advancements in MN technology, are platform-agnostic. This generality offers a unique template to interpret experimental observations, justify the recent introduction of drugs like GLP-1 cocktails, and optimize platforms for accurate and fast disease management. </p>

International Telemetering Conference Proceedings / October 17-20, 1994 / Town & Country Hotel and Conference Center, San Diego, California / Since the Gulf War, there has been significant interest in Theater Missile Defense (TMD) resulting in funding growth from tens of millions of dollars at the time of the Gulf War to $1.7 Billion in 1994. The Ballistic Missile Defense Organization (BMDO) has developed a Theater Missile Defense test and evaluation program that will assess technological feasibility and the degree to which system functionality and performance meet technical and operational requirements. The complexity of the TMD program necessitates a comprehensive test program which includes flight testing, ground testing, and modeling and simulation. This article will provide and overview the requirements and capabilities needed to satisfy these requirements. The data processing, and telemetry communities will play a major role in providing the expertise to support the development of the nation’s future Theater Missile Defense capabilities.

<p>The "Maker Movement" is a cultural phenomena rooted in DIY culture, which stresses making devices and creations on your own rather than purchasing it ready-made. At the core of the Maker Movement, is the "Maker Mindset"; a collection of attitudes, beliefs and behaviors that emphasize the importance of creativity, experimentation and innovation in the learning process. Since the Maker Mindset embodies constructionist principles at its core that push makers to experiment and problem-solve by collaborating with fellow makers through hands-on activities, it can be said that these activities comprise of Creative, Learning and Collaborative experiences. While Internet-of-Things devices have long been used to enhance these activities, research pertaining to using Augmented Reality in tandem with IoT for the purpose of enhancing experiences core to the Maker Mindset is relatively unexplored. Three different systems were developed with the goal of addressing this -- MicrokARts, ShARed IoT and MechARspace. Each system focuses on enhancing one of the three core experiences through AR-compatible IoT devices, whilst ensuring that they do not require prerequisite knowledge in order to author AR experiences. These systems were evaluated through user studies and testing over a variety of age-groups, with each system successfully enhancing one core experience each through the use of AR-IoT interactions.</p>