Abstract
There has been a growing interest in implementing complex machine learning algorithms such as convolutional neural networks (CNNs) on lower power embedded platforms to enable on-device learning and inference. Many of these platforms are to be deployed as low power sensor nodes with low to medium throughput requirement. Near threshold voltage (NTV) designs are well-suited for these applications but suffer from a significant increase in variations. In this paper, we propose a variation-tolerant architecture for CNNs capable of operating in NTV regime for energy efficiency. A statistical error compensation (SEC) technique referred to as rank decomposed SEC (RD-SEC) is proposed. The key idea is to exploit inherent redundancy within matrix-vector multiplication (or dot product ensemble), a power-hungry operation in CNNs, to derive low-cost estimators for error detection and compensation. When evaluated in CNNs for both the MNIST and CIFAR-10 datasets, simulation results in 45 nm CMOS show that RD-SEC enables robust CNNs operating in the NTV regime. Specifically, the proposed architecture can achieve up to 11 × improvement in variation tolerance and enable up to 113 × reduction in the standard deviation of detection accuracy Pdet while incurring marginal degradation in the median detection accuracy.
Original language | English (US) |
---|---|
Pages (from-to) | 1439-1451 |
Number of pages | 13 |
Journal | Journal of Signal Processing Systems |
Volume | 90 |
Issue number | 10 |
DOIs | |
State | Published - Oct 1 2018 |
Keywords
- Convolutional neural networks
- Near threshold voltage regime
- Rank decomposition
- Statistical error compensation
ASJC Scopus subject areas
- Control and Systems Engineering
- Theoretical Computer Science
- Signal Processing
- Information Systems
- Modeling and Simulation
- Hardware and Architecture