CNN2Gate: An Implementation of Convolutional Neural Networks Inference on FPGAs with Automated Design Space Exploration

Ghaffari, Alireza; Savaria, Yvon

doi:10.3390/electronics9122200

Cited by 23 publications

(11 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Proposed DNN achieved an accuracy rate of 99.6 percent, with a 1.75 percent increase over other similar projects with hardware implementation of the DNN using the XSG. The error rate implementation using (8,6) scenario was 7e-4 with a difference of 28 percent less than similar projects and using 306 fewer hardware components, which represents 30 percent of the FPGA device space. In conclusion, the proposed DNN design successfully reduced the hardware space utilization on FPGA devices while achieving a higher accuracy rate.…”

Section: Discussionmentioning

confidence: 92%

Reduced hardware requirements of deep neural network for breast cancer diagnosis

Tabra

Tawfeeq

2022

IJ-AI

View full text Add to dashboard Cite

Identifying breast cancer utilizing artificial intelligence technologies is valuable and has a great influence on the early detection of diseases. It also can save humanity by giving them a better chance to be treated in the earlier stages of cancer. During the last decade, deep neural networks (DNN) and machine learning (ML) systems have been widely used by almost every segment in medical centers due to their accurate identification and recognition of diseases, especially when trained using many datasets/samples. in this paper, a proposed two hidden layers DNN with a reduction in the number of additions and multiplications in each neuron. The number of bits and binary points of inputs and weights can be changed using the mask configuration on each subsystem to futher reduce the hardware requirements. The DNN was designed using a system generator and implemented using very hardware description language (VHDL). The system achievments outcomes the superior’s accuracy rate of approximately 99.6 percent in distinguishing bengin from malignant tissue. Also, the hardware resources were reduced by 30 percent from works of literature with an error rate of 7e-4 when using the Kintex-7 xc7k325t-3fbg676 board.

show abstract

Section: Discussionmentioning

confidence: 92%

Reduced hardware requirements of deep neural network for breast cancer diagnosis

Tabra

Tawfeeq

2022

IJ-AI

View full text Add to dashboard Cite

show abstract

“…The facilitation of the hardware implementation of deep-learning-based algorithms is an active research area. Recent efforts include fpga-ConvNet, Caffeine, and CNN2Gate that were developed by Venieris and Bouganis [111], Zhang et al [112], and Ghaffari and Savaria [113], respectively. These frameworks facilitated FPGA prototypes of deep neural networks that were designed using well-known libraries, such as PyTorch and Caffe.…”

Section: Real-time Processingmentioning

confidence: 99%

Visibility Restoration: A Systematic Review and Meta-Analysis

Ngo

Lee

Ngo

et al. 2021

Sensors

View full text Add to dashboard Cite

Image acquisition is a complex process that is affected by a wide variety of internal and environmental factors. Hence, visibility restoration is crucial for many high-level applications in photography and computer vision. This paper provides a systematic review and meta-analysis of visibility restoration algorithms with a focus on those that are pertinent to poor weather conditions. This paper starts with an introduction to optical image formation and then provides a comprehensive description of existing algorithms as well as a comparative evaluation. Subsequently, there is a thorough discussion on current difficulties that are worthy of a scientific effort. Moreover, this paper proposes a general framework for visibility restoration in hazy weather conditions while using haze-relevant features and maximum likelihood estimates. Finally, a discussion on the findings and future developments concludes this paper.

show abstract

“…In [12], the authors used reinforcement learning to explore optimization of deep neural networks on the ARM-Cortex-A CPUs. Likewise, in [13], the authors used a time-limited reinforcement learning [14] to execute design space exploration for deeply pipelined OpenCL kernels of the convolutional neural networks.…”

Section: Related Workmentioning

confidence: 99%

Efficient Design Space Exploration of OpenCL Kernels for FPGA Targets Using Black Box Optimization

Ghaffari

Savaria

2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

Nowadays, many industries are in favor of using intelligent design-space exploration as opposed to brute-force analysis. In many applications, the design-space is defined by multiple variables and their interactions. Although brute-force analysis is very simple, it is rarely scalable when the number of variables in the system increases. With the rising complexity of hardware designs, more intelligent approaches are needed to explore the design options. This paper proposes using smart meta-heuristic search algorithms such as Grey Wolf Optimization (GWO) in conjunction with Bayesian Optimization (BO) to solve this problem. We show that we can further reduce the design effort using a surrogate model that is created based on a novel hybrid GWO-BO method. The surrogate model is a useful abstraction to detect functional and physical inter-dependencies in the system in order to accurately predict its performance (e.g. throughput or latency). We evaluate our methodology and show that it can produce competitive results in order to find the best design variables that maximize performance of the system. Finally, we compare our results with previous statistical and heuristic methods proposed in the literature and find that the proposed GWO-BO method always performs better than the other considered methods.

show abstract

CNN2Gate: An Implementation of Convolutional Neural Networks Inference on FPGAs with Automated Design Space Exploration

Cited by 23 publications

References 29 publications

Reduced hardware requirements of deep neural network for breast cancer diagnosis

Reduced hardware requirements of deep neural network for breast cancer diagnosis

Visibility Restoration: A Systematic Review and Meta-Analysis

Efficient Design Space Exploration of OpenCL Kernels for FPGA Targets Using Black Box Optimization

Contact Info

Product

Resources

About