Fog computing (FC) is a distributed infrastructure computing that extends cloud computing (CC) capabilities to the edge of the network, closer to where data is generated and consumed. This approach responds to the cha...
详细信息
In this paper, a lossless audio codec is proposed by leveraging Wavelet transformation, Hierarchical encoding with Convolutional Neural Network architecture. In the first phase, three level 1D wavelet decomposition is...
详细信息
This paper adapts preexisting decision algorithms to a family RDF = {RDFa | n ∈ N} of languages regarding one-argument real functions;each RDFn is a quantifier-free theory about the differentiability class Cn, embody...
详细信息
In this paper, we proposed a genetic algorithm based on behavioral psychology developed by Carl Gustav Jung (16 Personalities model), in which we describe the person’s behavioral features related to his personality. ...
详细信息
The evolution of smart cities, driven by advancements in information transmission and the fusion of sensor technologies with IoT, faces challenges from substantial data traffic affecting Quality of Service factors lik...
详细信息
Deep neural networks have changed the current algorithms’ results in applications such as object classification, image segmentation or natural language processing. To increase their accuracy, they became more complex...
详细信息
Deep neural networks have changed the current algorithms’ results in applications such as object classification, image segmentation or natural language processing. To increase their accuracy, they became more complex and more costly in terms of storage, computation time and en-ergy consumption. This paper attacks the problem of storage and presents the advantages of using different number representations as fixed-point and posit numbers for deep neural network inference. The deep neural networks were trained using the proposed framework Low Precision Machine Learning (LPML) with 32-bit IEEE754. The storage was first optimized by the usage of knowledge distillation and then by modifying layer by layer the number representation together with the precision. The first significant results were made by modifying the number representation of the network but keeping the same precision per layer. For a 2-layer network (2LayerNet) using 16-bit Posit, the accuracy is 93.45%, close to 93.47%, the accuracy for using 32-bit IEEE754. Using the 8-bit Posit decreases the accuracy by 1.29%, but at the same time, it reduces the storage space by 75%. The usage of fixed point representation showed a small tolerance in the number of bits used for the fractional part. Using a 4-4 bit fixed point (4 bits for the integer part and 4 bits for the fractional part) reduces the storage used by 75% but decreases accuracy as low as 67.21%. When at least 8 bits are used for the fractional part, the results are similar to the 32-bit IEEE754. To increase accuracy before reducing precision, knowledge distillation was used. A ResNet18 network gained an 0.87% increase in accuracy by using a ResNet34 as a professor. By changing the number representation system and precision per layer, the storage was reduced by 43.47%, and the accuracy decreased by 0.26%. In conclusion, with the usage of knowledge distillation and change of number representation and precision per layer, the Resnet18 network had 66.75% sm
Scientific computing-based applications need a high load of computations. These computations are closely linked to the number representation system (NRS). A benchmark for decimal accuracy, storage space, computation t...
详细信息
The high volume and rapid pace of transactions generated by IoT devices pose challenges for current blockchain designs, which typically employ flat or two-tiered node organizations. These models often lack the scalabi...
详细信息
The importance of object detection within computer vision, especially in the context of detecting small objects, has notably increased. This thorough survey extensively examines small object detection across various a...
详细信息
Navigating the interaction landscape of Virtual Reality (VR) and Augmented Reality (AR) presents significant complexities due to the plethora of available input hardware and interaction modalities, compounded by spati...
详细信息
暂无评论