Javascript must be enabled to continue!
Quantization-Based Optimization Algorithm for Hardware Implementation of Convolution Neural Networks
View through CrossRef
Convolutional neural networks (CNNs) have demonstrated remarkable performance in many areas but require significant computation and storage resources. Quantization is an effective method to reduce CNN complexity and implementation. The main research objective is to develop a scalable quantization algorithm for CNN hardware design and model the performance metrics for the purpose of CNN implementation in resource-constrained devices (RCDs) and optimizing layers in deep neural networks (DNNs). The algorithm novelty is based on blending two quantization techniques to perform full model quantization with optimum accuracy, and without additional neurons. The algorithm is applied to a selected CNN model and implemented on an FPGA. Implementing CNN using broad data is not possible due to capacity issues. With the proposed quantization algorithm, we succeeded in implementing the model on the FPGA using 16-, 12-, and 8-bit quantization. Compared to the 16-bit design, the 8-bit design offers a 44% decrease in resource utilization, and achieves power and energy reductions of 41% and 42%, respectively. Models show that trading off one quantization bit yields savings of approximately 5.4K LUTs, 4% logic utilization, 46.9 mW power, and 147 μJ energy. The models were also used to estimate performance metrics for a sample DNN design.
Title: Quantization-Based Optimization Algorithm for Hardware Implementation of Convolution Neural Networks
Description:
Convolutional neural networks (CNNs) have demonstrated remarkable performance in many areas but require significant computation and storage resources.
Quantization is an effective method to reduce CNN complexity and implementation.
The main research objective is to develop a scalable quantization algorithm for CNN hardware design and model the performance metrics for the purpose of CNN implementation in resource-constrained devices (RCDs) and optimizing layers in deep neural networks (DNNs).
The algorithm novelty is based on blending two quantization techniques to perform full model quantization with optimum accuracy, and without additional neurons.
The algorithm is applied to a selected CNN model and implemented on an FPGA.
Implementing CNN using broad data is not possible due to capacity issues.
With the proposed quantization algorithm, we succeeded in implementing the model on the FPGA using 16-, 12-, and 8-bit quantization.
Compared to the 16-bit design, the 8-bit design offers a 44% decrease in resource utilization, and achieves power and energy reductions of 41% and 42%, respectively.
Models show that trading off one quantization bit yields savings of approximately 5.
4K LUTs, 4% logic utilization, 46.
9 mW power, and 147 μJ energy.
The models were also used to estimate performance metrics for a sample DNN design.
Related Results
Graph convolutional neural networks for 3D data analysis
Graph convolutional neural networks for 3D data analysis
(English) Deep Learning allows the extraction of complex features directly from raw input data, eliminating the need for hand-crafted features from the classical Machine Learning p...
SOC Reconfigurable Architecture for Software-Trained Neural Networks on FPGA
SOC Reconfigurable Architecture for Software-Trained Neural Networks on FPGA
Neural networks are extensively used in software and hardware applications. In hardware applications, it is necessary to
implement a small, accelerated, and configurable...
Constrained Quantization for Probability Distributions
Constrained Quantization for Probability Distributions
In this work, we extend the classical framework of quantization for Borel probability measures defined on normed spaces Rk by introducing and analyzing the notions of the nth const...
Silicon Micro-Robot With Neural Networks
Silicon Micro-Robot With Neural Networks
Insect type 4.0, 2.7, 2.5 mm. width, length, height size silicon micro-robot system with active hardware neural networks locomotion controlling system is presented in this chapter....
Fuzzy Chaotic Neural Networks
Fuzzy Chaotic Neural Networks
An understanding of the human brain’s local function has improved in recent years. But the cognition of human brain’s working process as a whole is still obscure. Both fuzzy logic ...
Research on Quantization Parameter Decision Scheme for High Efficiency Video Coding
Research on Quantization Parameter Decision Scheme for High Efficiency Video Coding
High-Efficiency Video Coding (HEVC) is one of the most widely studied coding standards. It still uses the block-based hybrid coding framework of Advanced Video Coding (AVC), and co...
Shrink and Eliminate: A Study of Post-Training Quantization and Repeated Operations Elimination in RNN Models
Shrink and Eliminate: A Study of Post-Training Quantization and Repeated Operations Elimination in RNN Models
Recurrent neural networks (RNNs) are neural networks (NN) designed for time-series applications. There is a growing interest in running RNNs to support these applications on edge d...
An Efficient Hardware Implementation of the Convolution Layer for the CNN Digit Recognition
An Efficient Hardware Implementation of the Convolution Layer for the CNN Digit Recognition
This work proposes an efficient hardware design of the convolution layer for a convolutional neural network (CNN) in digit recognition applications. CNN is a co...

