The design and construction of delivering announcements using a web-based SMS gateway aims to help overcome problems that often occur at pioneer airports;this is because when passengers want to travel, many passengers...
详细信息
Symbolic computation for systems of differential equations is often computationally expensive. Many practical differential models have a form of polynomial or rational OdE system with specified outputs. A basic symbol...
详细信息
deploying deep Neural Networks (dNNs) on edge devices to handle artificial intelligence (AI) tasks is increasingly important, but this is often limited by the computational and energy constraints of typical edge devic...
详细信息
ISBN:
(数字)9798350372601
ISBN:
(纸本)9798350372618
deploying deep Neural Networks (dNNs) on edge devices to handle artificial intelligence (AI) tasks is increasingly important, but this is often limited by the computational and energy constraints of typical edge devices. In this context, RISC-V-based internet of things edge computing engine, GAP8, provides a platform for edge AI. The engine features an 8-core cluster and ultra-low power consumption. Most importantly, GAP8 supports the deployment and acceleration of dNNs. However, there are limitations on memory size. In this research, we propose a hardware-adaptive optimization algorithm EN-HSPG to ensure high-speed operation of dNNs in edge AI. Specifically, the algorithm imposes sparsity-induced penalties to the weights and biases in batch normalization layers and selects layer-by-layer sparsity. Based on the optimizeddNN, we further explored the performance of GAP8 for dNNs. Experimental results show that after deploying dNNs, the parameter count and floating point operations pruning rate of the model optimized with the EN-HSPG algorithm reach as high as 98.82% and 91.28% respectively, using only 54% of L2 memory. Additionally, the inference time of the EN-HSPG optimized model is just 115.26ms on the GAP8. Thus, the research enables the operation of complex deep learning networks on edge devices with limited computing capabilities.
Cytometry plays a crucial role in characterizing cell properties, but its restricted optical window (400-850 nm) limits the number of stained fluorophores that can be detected simultaneously and hampers the study and ...
详细信息
Students’ computational thinking andprogramming skills may grow due to collaborative programming. But as the researchers have noted, students frequently do not use metacognition to manage their cognitive activities ...
详细信息
An immobilized hydrogen-producing consortium investigated biohydrogen production from food waste using a combination of sodium alginate and cassava rhizome biochar. We investigated the effect of varying the biochar co...
详细信息
A uniformly oriented purple membrane (PM) monolayer containing photoactive bacteriorhodopsin has recently been applied as a sensitive photoelectric transducer to assay color proteins and microbes quantitatively. This ...
详细信息
This paper presents an intelligent optimization of the AI/ML framework to detect network anomalies via time collection analysis. The proposed framework entails leveraging an AI / ML version to manner the time series f...
详细信息
AC-dC converters almost always require common-mode (CM) filters to restrict conducting their electromagnetic interference (EMI) emissions to the power grid and ensure compliance with EMI certification requirements. Th...
详细信息
ISBN:
(数字)9784885523472
ISBN:
(纸本)9798350349498
AC-dC converters almost always require common-mode (CM) filters to restrict conducting their electromagnetic interference (EMI) emissions to the power grid and ensure compliance with EMI certification requirements. These CM filters, which are based on the well-known LC lowpass filter, are typically bulky and expensive. Hence, a method for passive $\mathbf{C M}$ current cancellation using a nonlinear model is proposed herein. This method is entirely composed of passive components and offers advantages in terms of cost and system reliability. Compared to conventional passive LC filters, the number of magnetic components remains unchanged. This study proposes the derivation of a nonlinear model to estimate the filter performance using measureddata from an actual CM choke; however, the proposed methoddoes not require the mathematical model of a CM choke.
Wheelchair basketball for people with disabilities is a niche market with limited attention and resources due to its smaller scale. We utilize the characteristics of VR technology, combining wearable devices with phys...
详细信息
暂无评论