Modern field-programmable Gate Arrays (FPGAs) are highly versatile, with reconfigurable logic functionality that allows designers to create custom designs. Unlike traditional fixed-function integrated circuits, FPGAs ...
详细信息
ISBN:
(数字)9798331522445
ISBN:
(纸本)9798331522452
Modern field-programmable Gate Arrays (FPGAs) are highly versatile, with reconfigurable logic functionality that allows designers to create custom designs. Unlike traditional fixed-function integrated circuits, FPGAs are reconfigured multiple times to perform different tasks, making them ideal for various applications. An Embedded FPGA (eFPGA) takes the concept of an FPGA and integrates it as an IP core within a larger System-on-Chip (SoC) or Application-Specific Integrated Circuit (ASIC). the proposed work employs Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to evolve efficient and high-performing eFPGA architectures with reduced critical path delay and power defined by a sufficient number of Configurable logic Blocks (CLBs), custom Digital Signal Processors (DSPs) and Block RAM (BRAMs) for targetting workload under consideration. As a result, this work presents a well-balanced eFPGA architecture layout obtained for four workloads equipped with different numbers of heterogeneous and configurable logic blocks. the results showed a significant improvement of 44.62% in hardware parameters.
the HLS toolchain effectively reduces the design complexity of FPGA hardware accelerators. However, in scenarios involving the multi-objective optimization of large-scale HLS designs, determining the knob configuratio...
详细信息
ISBN:
(数字)9798331530075
ISBN:
(纸本)9798331530082
the HLS toolchain effectively reduces the design complexity of FPGA hardware accelerators. However, in scenarios involving the multi-objective optimization of large-scale HLS designs, determining the knob configurations of Pareto design points remains a challenging task for designers. Our work re-evaluates the key factors affecting the efficiency of multiobjective design space exploration in HLS design and proposes an efficient framework named FlexWalker. It utilizes the upper confidence bound algorithm to organize various heterogeneous regression models for predicting the quality of HLS designs with different knob configurations in the design space and introduces a probability sampling algorithm and an elastic Pareto frontier to counteract the negative impact of regression model errors. Experimental results show that our work can stably eliminate over 90% of non-Pareto frontier design points in the tested HLS design space, effectively enhancing the efficiency of multiobjective design space exploration.
the proceedings contain 18 papers. the topics discussed include: angular responsivity of ground and space-based direct solar irradiance radiometers;stray-light correction methodology for the precision solar spectrorad...
the proceedings contain 18 papers. the topics discussed include: angular responsivity of ground and space-based direct solar irradiance radiometers;stray-light correction methodology for the precision solar spectroradiometer;a simple method of UV stray light correction for field spectrometers in ground validation sites;advancements in NRC's primary spectral irradiance scale realization;calibration of a modular trap detector system towards a new realization of the luminous intensity unit;novel LED-based radiation source and its application in diffuse reflectometry and polarization measurements;and characterization of bidirectional transmissive and reflective properties of black silicon.
the aim of this work is to develop a video processing system with efficient hardware resource using a fieldprogrammable gate array (FPGA). the FPGA board contains several resources that may be implemented on both har...
详细信息
Although analyzing high-stakes medical images is challenging, it is what recent deep neural networks quite excel at. Noticing that there are still many concerns regarding the adoption of vision AI-based digital human ...
详细信息
ISBN:
(纸本)9783031357473;9783031357480
Although analyzing high-stakes medical images is challenging, it is what recent deep neural networks quite excel at. Noticing that there are still many concerns regarding the adoption of vision AI-based digital human models in actual medical practices, this paper goes through challenges that occur in each development stage of vision AI in medical applications. Focusing on modeling patients' internal organisms via medical imaging, we found that most existing vision-based AI systems share similar challenges, ranging from huge computational resources, laborious data annotation, domain shifting, and unexplainability. Data collection in this era of data privacy law is another challenge that is rarely discussed in previous works conducting technical implementations of deep neural networks. In the end, our conclusion is that leading researchers in deep neural networks tend to put more concern into introducing techniques that allow newer, bigger, and more precise networks to be easily trained and evaluated based on some quantitative evaluation metrics. Meanwhile, physicians and data protection laws seem to hold a different concern regarding qualitative issues about how to make these deep networks trustworthy, ethical, and able to explain their decisions as well as underneathlogic in a meaningful manner.
High-performance implementation of non-linear support vector machine (SVM) function is important in many applications. this paper develops a hardware design of Gaussian kernel function with high-performance since it i...
详细信息
Multi-Agent Reinforcement Learning (MARL) is an emerging technology that has seen success in many AI applications. Multi-Actor-Attention-Critic (MAAC) is a state-of-the-art MARL algorithm that uses a Multi-Head Attent...
详细信息
ISBN:
(数字)9798331530075
ISBN:
(纸本)9798331530082
Multi-Agent Reinforcement Learning (MARL) is an emerging technology that has seen success in many AI applications. Multi-Actor-Attention-Critic (MAAC) is a state-of-the-art MARL algorithm that uses a Multi-Head Attention (MHA) mechanism to learn messages communicated among agents during the training process. Current implementations of MAAC using CPU and CPU-GPU platforms lack fine-grained parallelism among agents, sequentially executing each stage of the training loop, and their performance suffers from costly data movement involved in MHA communication learning. In this work, we develop the first high-throughput accelerator for MARL with attention-based communication on a CPU-FPGA heterogeneous system. We alleviate the limitations of existing implementations through a combination of data- and pipeline-parallel modules in our accelerator design and enable fine-grained system scheduling for exploiting concurrency among heterogeneous resources. Our design increases the overall system throughput by $4.6 \times$ and $4.1 \times$ compared to CPU and CPU-GPU implementations, respectively.
Scientific Machine Reading Comprehension (SMRC) aims to facilitate the understanding of scientific texts through human-machine interactions. While existing dataset has significantly contributed to this field, it predo...
详细信息
To summarize the research and progress in the preparation of anthocyanin indicator labels and their application in food freshness monitoring. Relevant domestic and international literature is reviewed and organized, a...
详细信息
the field R of real numbers is obtained from the rational numbers Q by taking the completion with respect to the usual absolute value. We then define the complex numbers C as an algebraic closure of R. the p-adic anal...
详细信息
暂无评论