Space Efficient Quantization for Deep Convolutional Neural Networks

Deep convolutional neural networks (DCNNs) have shown outstanding performance in the fields of computer vision, natural language processing, and complex system analysis. With the improvement of performance with deeper layers, DCNNs incur higher computational complexity and larger storage requirement, making it extremely difficult to deploy DCNNs on resource-limited embedded systems (such as mobile devices or Internet of Things devices). Network quantization efficiently reduces storage space required by DCNNs. However, the performance of DCNNs often drops rapidly as the quantization bit reduces. In this article, we propose a space efficient quantization scheme which uses eight or less bits to represent the original 32-bit weights. We adopt singular value decomposition (SVD) method to decrease the parameter size of fully-connected layers for further compression. Additionally, we propose a weight clipping method based on dynamic boundary to improve the performance when using lower precision. Experimental results demonstrate that our approach can achieve up to approximately 14x compression while preserving almost the same accuracy compared with the full-precision models. The proposed weight clipping method can also significantly improve the performance of DCNNs when lower precision is required.

     

Inorganic thermoelectric materials: A review

Thermoelectric generator, which converts heat into electrical energy, has great potential to power portable devices. Nevertheless, the efficiency of a thermoelectric generator suffers due to inefficient thermoelectric material performance. In the last two decades, the performance of inorganic thermoelectric materials has been significantly advanced through rigorous efforts and novel techniques. In this review, major issues and recent advancements that are associated with the efficiency of inorganic thermoelectric materials are encapsulated. In addition, miscellaneous optimization strategies, such as band engineering, energy filtering, modulation doping, and low dimensional materials to improve the performance of inorganic thermoelectric materials are reported. The methodological reviews and analyses showed that all these techniques have significantly enhanced the Seebeck coefficient, electrical conductivity, and reduced the thermal conductivity, consequently, improved ZT value to 2.42, 2.6, and 1.85 for near-room, medium, and high temperature inorganic thermoelectric material, respectively. Moreover, this review also focuses on the performance of silicon nanowires and their common fabrication techniques, which have the potential for thermoelectric power generation. Finally, the key outcomes along with future directions from this review are discussed at the end of this article.     

Heat transfer analysis for particle–fluid suspension thermomagnetohydrodynamic peristaltic flow with Darcy–Forchheimer medium

This theoretical analysis explores the effect of heat and mass transfer on particle–fluid suspension for the Rabinowitsch fluid model with the stiffness and dynamic damping effects through Darcy–Brinkman–Forchheimer porous medium. In this study, we also incorporate slip and transverse magnetic field effects. Using low Reynolds number, to neglect inertial forces and to keep the pressure constant during the flow, channel height is used largely as compared with the ratio of length of the wave. A numerical technique is used to solve flow governing system of differential equations. Particular attention is paid to viscous damping force parameter, stiffness parameter, and rigidity parameter; also, the numerical data for thermal profile, momentum, and concentration distribution are presented graphically. Outcomes are deliberated in detail for different fluid models (thinning, thickening, and viscous models). It is found that velocity profile increases for greater values of viscous damping effect and stiffness and rigidity parameter for shear thinning, but conflicting comportment is showed for thickening nature model. Viscous dissipation effects increases the thermal profile for all cases of fluid models. The scope of the present article is valuable in explaining the blood transport dynamics in small vessels while considering the important wall features with chemical reaction characteristics. The current analysis has extensive applications in biomedical engineering field, that is, peristaltic pumps.     

Conditional Preference Networks with User's Genuine Decisions

User's choices involve habitual behavior and genuine decision. Habitual behavior is often expressed using preferences. In a multiattribute case, the Conditional Preference Network (CP-net) is a graphical model to represent user's conditional ceteris paribus (all else being equal) preference statements. Indeed, the CP-net induces a strict partial order over the outcomes. By contrast, we argue that genuine decisions are environmentally influenced and introduce the notion of “comfort” to represent this type of choices. In this article, we propose an extension of the CP-net model that we call the CP-net with Comfort (CPC-net) to represent a user's comfort with preferences. Given that preference and comfort might be two conflicting objectives, we define the Pareto optimality of outcomes when achieving outcome optimization with respect to a given CPC-net. Then, we propose a backtrack search algorithm to find the Pareto optimal outcomes. On the other hand, two outcomes can stand in one of six possible relations with respect to a CPC-net. The exact relation can be obtained by performing dominance testing in the corresponding CP-net and comparing the numeric comforts.     

MRMR Based Feature Vector Design for Efficient Citrus Disease Detection

In recent times, the images and videos have emerged as one of the most important information source depicting the real time scenarios. Digital images nowadays serve as input for many applications and replacing the manual methods due to their capabilities of 3D scene representation in 2D plane. The capabilities of digital images along with utilization of machine learning methodologies are showing promising accuracies in many applications of prediction and pattern recognition. One of the application fields pertains to detection of diseases occurring in the plants, which are destroying the widespread fields. Traditionally the disease detection process was done by a domain expert using manual examination and laboratory tests. This is a tedious and time consuming process and does not suffice the accuracy levels. This creates a room for the research in developing automation based methods where the images captured through sensors and cameras will be used for detection of disease and control its spreading. The digital images captured from the field's forms the dataset which trains the machine learning models to predict the nature of the disease. The accuracy of these models is greatly affected by the amount of noise and ailments present in the input images, appropriate segmentation methodology, feature vector development and the choice of machine learning algorithm. To ensure the high rated performance of the designed system the research is moving in a direction to fine tune each and every stage separately considering their dependencies on subsequent stages. Therefore the most optimum solution can be obtained by considering the image processing methodologies for improving the quality of image and then applying statistical methods for feature extraction and selection. The training vector thus developed is capable of presenting the relationship between the feature values and the target class. In this article, a highly accurate system model for detecting the diseases occurring in citrus fruits using a hybrid feature development approach is proposed. The overall improvement in terms of accuracy is measured and depicted.     

Protein–Energy Malnutrition: A Risk Factor for Various Ailments

The wheel of industrialization that spun throughout the last century resulted in urbanization coupled with modifications in lifestyles and dietary habits. However, the communities living in developing economies are facing many problems related to their diet and health. Amongst, the prevalence of nutritional problems especially protein–energy malnutrition (PEM) and micronutrients deficiencies are the rising issues. Moreover, the immunity or susceptibility to infect-parasitic diseases is also directly linked with the nutritional status of the host. Likewise, disease-related malnutrition that includes an inflammatory component is commonly observed in clinical practice thus affecting the quality of life. The PEM is treatable but early detection is a key for its appropriate management. However, controlling the menace of PEM requires an aggressive partnership between the physician and the dietitian. This review mainly attempts to describe the pathophysiology, prevalence and consequences of PEM and aims to highlight the importance of this clinical syndrome and the recent growth in our understanding of the processes behind its development. Some management strategies/remedies to overcome PEM are also the limelight of the article. In the nutshell, early recognition, prompt management, and robust follow up are critical for best outcomes in preventing and treating PEM.