ABSTRACT
Memristor-based synaptic network has been widely investigated and applied to neuromorphic computing systems for the fast computation and low design cost. As memristors continue to mature and achieve higher density, bit failures within crossbar arrays can become a critical issue. These can degrade the computation accuracy significantly. In this work, we propose a defect rescuing design to restore the computation accuracy. In our proposed design, significant weights in a specified network are first identified and retraining and remapping algorithms are described. For a two layer neural network with 92.64% classification accuracy on MNIST digit recognition, our evaluation based on real device testing shows that our design can recover almost its full performance when 20% random defects are present.
- P. Simon, Too Big to Ignore: The Business Case for Big Data. Wiley, 2013. Google ScholarDigital Library
- D. C. Ciresan et al., "Convolutional Neural Network Committees for Handwritten Character Classification," in 2011 International Conference on Document Analysis and Recognition, pp. 1135--1139, Sept 2011. Google ScholarDigital Library
- V. Vanhoucke et al., "Improving the speed of neural networks on CPUs," in Deep Learning and Unsupervised Feature Learning Workshop, 2011.Google Scholar
- C. Farabet et al., Large-Scale FPGA-based Convolutional Networks. Cambridge university Press, 2011.Google Scholar
- W. A. Wulf and S. A. McKee, "Hitting the Memory Wall: Implications of the Obvious," ACM SIGARCH computer architecture news, vol. 23, no. 1, pp. 20--24, 1995. Google ScholarDigital Library
- P. Merolla et al., "A digital neurosynaptic core using embedded crossbar memory with 45pJ per spike in 45nm," in CICC, pp. 1--4, 2011.Google Scholar
- P. A. Merolla et al., "A million spiking-neuron integrated circuit with a scalable communication network and interface," Science, vol. 345, no. 6197, pp. 668--673, 2014.Google ScholarCross Ref
- M. Sharad et al., "Proposal For Neuromorphic Hardware Using Spin Devices," in Disordered Systems and Neural Networks, 2012.Google Scholar
- C. Liu i "A Spiking Neuromorphic Design with Resistive Crossbar," in DAC, pp. 1--6, July 2015. Google ScholarDigital Library
- C. Liu et al., "A Memristor Crossbar Based Computing Enginee Optimized for High Speed and Accuracy," in ISVLSI, July 2016.Google Scholar
- M. Prezioso et al., "Training and operation of an integrated neuromorphic network based on metal-oxide memristors," Nature, vol. 521, pp. 61--64, 2015.Google ScholarCross Ref
- M. Hu et al., "Dot-Product Engine for Neuromorphic Computing: Programming 1T1M Crossbar to Accelerate Matrix-Vector Multiplication," in DAC, pp. 1--6, 2016. Google ScholarDigital Library
- L. O. Chua, "Memristor-the missing circuit element," in IEEE Transactions on Circuit Theory, vol. 18, pp. 507--519, September 1971.Google ScholarCross Ref
- D. B. Strukov et al., "The Missing Memristor Found," Nature, vol. 453, no. 12, pp. 28--35, 2008.Google Scholar
- F. M. Bayat, M. Prezioso, B. Chakrabarti, I. Kataeva, and D. Strukov, "Advancing memristive analog neuromorphic networks: Increasing complexity, and coping with imperfect hardware components," arXiv preprint arXiv:1611.04465, 2016.Google Scholar
- S. Kannan et al., "Detection, Diagnosis, and Repair of Faults in Memristor-based Memories," in VTS, pp. 1--6, April 2014.Google Scholar
- E. J. Merced-Grafals et al., "Repeatable, Accurate, and High Speed Multi-level Programming of Memristor 1T1R Arrays for Power Efficient Analog Computing Applications," Nanotechnology, vol. 27, no. 36, p. 365202, 2016.Google ScholarCross Ref
- N. Srivastava et al., "Dropout: A Simple Way to Prevent Neural Networks from Overfitting," Journal of Machine Learning Research, vol. 15, pp. 1929--1958, 2014. Google ScholarDigital Library
- S. Han et al., "Learning both Weights and Connections for Efficient Neural Network," in Advances in Neural Information Processing Systems, pp. 1135--1143, 2015. Google ScholarDigital Library
- S. Han et al., "EIE: Efficient Inference Engine on Compressed Deep Neural Network," in ISCA, pp. 243--254, 2016. Google ScholarDigital Library
- S. Han et al., "DSD: Regularizing Deep Neural Networks with Dense-Sparse-Dense Training Flow," arXiv preprint arXiv:1607.04381, 2016.Google Scholar
- T. M. Mitchell, Machine Learning. McGraw-Hill Science, 1997. Google ScholarDigital Library
- Y. LeCun et al., "The MNIST database of handwritten digits." http://yann.lecun.com/exdb/mnist/.Google Scholar
- I. Kataeva, F. Merrikh-Bayat, E. Zamanidoost, and D. Strukov, "Efficient training algorithms for neural networks based on memristive crossbar circuits," in Neural Networks (IJCNN), 2015 International Joint Conference on, pp. 1--8, IEEE, 2015.Google Scholar
- Rescuing Memristor-based Neuromorphic Design with High Defects
Recommendations
Learning Behavior of Memristor-Based Neuromorphic Circuits in the Presence of Radiation
ICONS '19: Proceedings of the International Conference on Neuromorphic SystemsIn this paper, a feed-forward spiking neural network with memristive synapses is designed to learn a spatio-temporal pattern representing the 25-pixel character 'B' by separating correlated and uncorrelated afferents. The network uses spike-timing-...
Neuromorphic Computing Based on Memristor
ICASIT 2020: Proceedings of the 2020 International Conference on Aviation Safety and Information TechnologyBioelectronics is defined as a combination of electronic principles and biological phenomena. While many major fields such as Medical treatment, technology advancement, healthcare are increasing its demand for bioelectronic circuits. And the ...
Sequential Memristor Crossbar for Neuromorphic Pattern Recognition
Most of human's intelligent behaviors such as inference, prediction, anticipation, etc. are based on the processing of sequential data from human's sensory systems. Thus, a sequential memory that can process sequential information is very essential to ...
Comments