Using Libraries of Approximate Circuits in Design of Hardware Accelerators of Deep Neural Networks

04/22/2020
by   Vojtech Mrazek, et al.
0

Approximate circuits have been developed to provide good tradeoffs between power consumption and quality of service in error resilient applications such as hardware accelerators of deep neural networks (DNN). In order to accelerate the approximate circuit design process and to support a fair benchmarking of circuit approximation methods, libraries of approximate circuits have been introduced. For example, EvoApprox8b contains hundreds of 8-bit approximate adders and multipliers. By means of genetic programming we generated an extended version of the library in which thousands of 8- to 128-bit approximate arithmetic circuits are included. These circuits form Pareto fronts with respect to several error metrics, power consumption and other circuit parameters. In our case study we show how a large set of approximate multipliers can be used to perform a resilience analysis of a hardware accelerator of ResNet DNN and to select the most suitable approximate multiplier for a given application. Results are reported for various instances of the ResNet DNN trained on CIFAR-10 benchmark problem.

READ FULL TEXT
research
03/09/2022

ArithsGen: Arithmetic Circuit Generator for Hardware Accelerators

Generators of arithmetic circuits can automatically deliver various impl...
research
02/22/2019

autoAx: An Automatic Design Space Exploration and Circuit Building Methodology utilizing Libraries of Approximate Components

Approximate computing is an emerging paradigm for developing highly ener...
research
02/21/2020

TFApprox: Towards a Fast Emulation of DNN Approximate Hardware Accelerators on GPU

Energy efficiency of hardware accelerators of deep neural networks (DNN)...
research
03/11/2019

Automated Circuit Approximation Method Driven by Data Distribution

We propose an application-tailored data-driven fully automated method fo...
research
06/27/2022

Designing Approximate Arithmetic Circuits with Combined Error Constraints

Approximate circuits trading the power consumption for the quality of re...
research
07/01/2022

RUCA: RUntime Configurable Approximate Circuits with Self-Correcting Capability

Approximate computing is an emerging computing paradigm that offers impr...
research
04/22/2020

ApproxFPGAs: Embracing ASIC-Based Approximate Arithmetic Components for FPGA-Based Systems

There has been abundant research on the development of Approximate Circu...

Please sign up or login with your details

Forgot password? Click here to reset