CHARM: Composing Heterogeneous AcceleRators for Matrix Multiply on Versal ACAP Architecture [PDF]
Symposium on Field Programmable Gate Arrays, 2023 Dense matrix multiply (MM) serves as one of the most heavily used kernels in deep learning applications. To cope with the high computation demands of these applications, heterogeneous architectures featuring both FPGA and dedicated ASIC accelerators have
Jinming Zhuang +12 more
semanticscholar +1 more source
A study of electronic and transport properties of CsSnBr3: A first principle study [PDF]
مجله علوم و فنون هستهای, 2023 CsSnBr3 nanocrystals are better than other lead-free perovskites because of their ease and low-cost synthesis, long-term function, and good stability. It is a suitable selection for use in tandem photodetectors.
S. Nazari +3 more
doaj +1 more source
Analysis of Irradiation Induced Defect Clusterization for Zr-1%Nb Alloy Using Atomistic Simulation [PDF]
Journal of Nuclear Research and Applications, 2023 Nuclear-grade zirconium alloys properties are very similar to those of pure zirconium (Zr), because in most cases they contain more than 95% of Zr atoms. They have extensive application in nuclear industry, especially in fuel cladding. Lattice properties
M. R. Basaadat, M. Payami, S. Sheykhi
doaj +1 more source
Hardware Approximate Techniques for Deep Neural Network Accelerators: A Survey [PDF]
ACM Computing Surveys, 2022 Deep Neural Networks (DNNs) are very popular because of their high performance in various cognitive tasks in Machine Learning (ML). Recent advancements in DNNs have brought levels beyond human accuracy in many tasks, but at the cost of high computational
Giorgos Armeniakos +3 more
semanticscholar +1 more source
Favorable experimental conditions for differential cross-section measurement of PIGE reactions using the van de graaff accelerator of tehran [PDF]
مجله علوم و فنون هستهای, 2021 The present research aims to measure the physical parameters affecting the differential cross-sections of PIGE reactions in the 45˚R beamline of the Van de Graaff accelerator.
A. Jokar, O. Kakuee, M. Lamehi-Rachti
doaj +1 more source
CoSA: Scheduling by Constrained Optimization for Spatial Accelerators [PDF]
International Symposium on Computer Architecture, 2021 Recent advances in Deep Neural Networks (DNNs) have led to active development of specialized DNN accelerators, many of which feature a large number of processing elements laid out spatially, together with a multi-level memory hierarchy and flexible ...
Qijing Huang +7 more
semanticscholar +1 more source
Applied Sciences, 2021 The electro-optical process is a popular method for terahertz radiation detection. Detectors based on the electro-optical process have large bandwidth, and the signal-to-noise ratio (SNR) is relatively high.
Adnan Haj Yahya +4 more
doaj +1 more source
An Evaluation of Edge TPU Accelerators for Convolutional Neural Networks [PDF]
IEEE International Symposium on Workload Characterization, 2021 Edge TPUs are a domain of accelerators for low-power, edge devices and are widely used in various Google products such as Coral and Pixel devices. In this paper, we first discuss the major microarchitectural details of Edge TPUs.
A. Yazdanbakhsh +4 more
semanticscholar +1 more source
Inter-layer Scheduling Space Definition and Exploration for Tiled Accelerators
International Symposium on Computer Architecture, 2023 With the continuous expansion of the DNN accelerator scale, inter-layer scheduling, which studies the allocation of computing resources to each layer and the computing order of all layers in a DNN, plays an increasingly important role in maintaining a ...
Jingwei Cai +4 more
semanticscholar +1 more source