ABSTRACT
Convolutional Neural Network (CNN) is a powerful technique widely used in computer vision area, which also demands much more computations and memory resources than traditional solutions. The emerging metal-oxide resistive random-access memory (RRAM) and RRAM crossbar have shown great potential on neuromorphic applications with high energy efficiency. However, the interfaces between analog RRAM crossbars and digital peripheral functions, namely Analog-to-Digital Converters (ADCs) and Digital-to-Analog Converters (DACs), consume most of the area and energy of RRAM-based CNN design due to the large amount of intermediate data in CNN. In this paper, we propose an energy efficient structure for RRAM-based CNN. Based on the analysis of data distribution, a quantization method is proposed to transfer the intermediate data into 1 bit and eliminate DACs. An energy efficient structure using input data as selection signals is proposed to reduce the ADC cost for merging results of multiple crossbars. The experimental results show that the proposed method and structure can save 80% area and more than 95% energy while maintaining the same or comparable classification accuracy of CNN on MNIST.
- J. Qiu et al., "Going deeper with embedded fpga platform for convolutional neural network," in ACM International Symposium on FPGA, 2016. Google ScholarDigital Library
- C. Zhang et al., "Optimizing fpga-based accelerator design for deep convolutional neural networks," in FPGA, 2015, pp. 161--170. Google ScholarDigital Library
- M. Hu et al., "Hardware realization of bsb recall function using memristor crossbar arrays," in DAC, 2012, pp. Google ScholarDigital Library
- B. Li et al., "Memristor-based approximated computation," in ISLPED. IEEE Press, 2013, pp. 242--247. Google ScholarDigital Library
- C. Xu et al., "Design implications of memristor-based rram cross-point structures," in DATE, 2011, pp. 1--6.Google Scholar
- L. Xia et al., "MNSIM: Simulation platform for memristor-based neuromorphic computing system," in DATE, 2016, pp. 469--474.Google ScholarCross Ref
- K. Simonyan et al., "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.Google Scholar
- Y. Zhang et al., "Random telegraph noise analysis in alox/woy resistive switching memories," Applied Physics Letters, vol. 104, no. 10, p. 103507, 2014.Google ScholarCross Ref
- Y. LeCun et al., "The mnist database of handwritten digits," 1998.Google Scholar
- M. Kim et al., "Bitwise neural networks," in ICML workshop, 2015.Google Scholar
- J. Fieres et al., "Training convolutional networks of threshold neurons suited for low-power hardware implementation," in IJCNN. IEEE, 2006, pp. 21--28.Google Scholar
- A. Krizhevsky et al., "Imagenet classification with deep convolutional neural networks," in NIPS, 2012, pp. 1097--1105.Google ScholarDigital Library
- F. Alibart et al., "High precision tuning of state for memristive devices by adaptable variation-tolerant algorithm," Nanotechnology, vol. 23, no. 7, p. 075201, 2012.Google ScholarCross Ref
- D. Garbin et al., "Hfo2-based oxram devices as synapses for convolutional neural networks," TED, vol. 62, no. 8, pp. 2494--2501, Aug 2015.Google ScholarCross Ref
- M. Catanzaro et al., "Reconfigurable rram for lut logic mapping: A case study for reliability enhancement," in SOCC, 2012, pp. 94--99.Google Scholar
- Y. Zhang et al., "Study of conduction and switching mechanisms in Al/AlOx/WOx/W resistive switching memory for multilevel applications," Applied Physics Letters, vol. 102, no. 23, p. 233502, 2013.Google ScholarCross Ref
- R. St Amant et al., "General-purpose code acceleration with limited-precision analog computation," in ISCA, 2014, pp. 505--516. Google ScholarDigital Library
- W.-H. Tseng et al., "A 960ms/s dac with 80db sfdr in 20nm cmos for multi-mode baseband wireless transmitter," in VLSI Circuits Digest of Technical Papers, 2014, pp. 1--2.Google Scholar
- B. Li et al., "Merging the interface: Power, area and accuracy co-optimization for rram crossbar-based mixed-signal computing system," in DAC, 2015. Google ScholarDigital Library
- S. Han et al., "Learning both weights and connections for efficient neural networks," arXiv preprint arXiv:1506.02626, 2015.Google Scholar
- S. Yu et al., "A low energy oxide-based electronic synaptic device for neuromorphic visual systems with tolerance to device variation," Advanced Materials, vol. 25, no. 12, pp. 1774--1779, 2013.Google ScholarCross Ref
- T. Tang et al., "Spiking neural network with rram: Can we use it for real-world application?" in DATE, 2015, pp. 860--865. Google ScholarDigital Library
Recommendations
A fully differential comparator-based switched-capacitor ΔΣ modulator
In this brief, a fully differential comparator-based switched-capacitor (CBSC) second-order delta-sigma (ΔΣ) modulator is Presented. To ensure differential operation, the CBSC ΔΣ modulator utilizes a common-mode feedback circuit to balance the pull-up ...
A differential switched-capacitor amplifier with programmable gain and output offset voltage
SBCCI '06: Proceedings of the 19th annual symposium on Integrated circuits and systems designThe design of a low-power differential Switched-Capacitor (SC) amplifier for processing a fully-differential input signal coming from a pressure sensor interface is reported. The circuit is intended to amplify the input signal, convert it to single ...
Cascode Configurations for Switched Current Copiers
Special issue: papers from the 40th midwest symposiumVarious cascode circuits are investigated with regard to their suitability for switched current copier applications. A generalized circuit representation for cascodes is introduced and different cascode circuits are compared with respect to their small-...
Comments