8 October 2019 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Improving the accuracy of a neural network (NN) usually requires using larger hardware that consumes more energy. However, the error tolerance of NNs and their applications allow approximate computing techniques to be applied to reduce implementation costs. Given that multiplication is the most resource-intensive and power-hungry operation in NNs, more economical approximate multipliers (AMs) can significantly reduce hardware costs. In this article, we show that using AMs can also improve the NN accuracy by introducing noise. We consider two categories of AMs: 1) deliberately designed and 2) Cartesian genetic programing (CGP)-based AMs. The exact multipliers in two representative NNs, a multilayer perceptron (MLP) and a convolutional NN (CNN), are replaced with approximate designs to evaluate their effect on the classification accuracy of the Mixed National Institute of Standards and Technology (MNIST) and Street View House Numbers (SVHN) data sets, respectively. Interestingly, up to 0.63% improvement in the classification accuracy is achieved with reductions of 71.45% and 61.55% in the energy consumption and area, respectively. Finally, the features in an AM are identified that tend to make one design outperform others with respect to NN accuracy. Those features are then used to train a predictor that indicates how well an AM is likely to work in an NN.
8 July 2014 in IEEE Transactions on Evolutionary Computation
In approximate computing, the requirement of perfect functional behavior can be relaxed because some applications are inherently error resilient. Approximate circuits, which fall into the approximate computing paradigm, are designed in such a way that they do not fully implement the logic behavior given by the specification and, hence, their accuracy can be exchanged for lower area, delay or power consumption. In order to automate the design process, we propose to evolve approximate digital circuits that show a minimal error for a supplied amount of resources. The design process, which is based on Cartesian genetic programming (CGP), can be repeated many times in order to obtain various tradeoffs between the accuracy and area. A heuristic seeding mechanism is introduced to CGP, which allows for improving not only the quality of evolved circuits, but also reducing the time of evolution. The efficiency of the proposed method is evaluated for the gate as well as the functional level evolution. In particular, approximate multipliers and median circuits that show very good parameters in comparison with other available implementations were constructed by means of the proposed method.
20 October 2020 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Libraries of approximate circuits are composed of fully characterized digital circuits that can be used as building blocks of energy-efficient implementations of hardware accelerators. They can be employed not only to speed up the accelerator development but also to analyze how an accelerator responds to introducing various approximate operations. In this paper, we present a methodology that automatically builds comprehensive libraries of approximate circuits with desired properties. Target approximate circuits are generated using Cartesian genetic programming. In addition to extending the EvoApprox8b library that contains common approximate arithmetic circuits, we show how to generate more specific approximate circuits; in particular, MxN-bit approximate multipliers that exhibit promising results when deployed in convolutional neural networks. By means of the evolved approximate multipliers, we perform a detailed error resilience analysis of five different ResNet networks. We identify the convolutional layers that are good candidates for adopting the approximate multipliers and suggest particular approximate multipliers whose application can lead to the best trade-offs between the classification accuracy and energy requirements. Experiments are reported for CIFAR-10 and CIFAR-100 data sets.