Journal
NATURE COMMUNICATIONS
Volume 12, Issue 1, Pages -Publisher
NATURE PORTFOLIO
DOI: 10.1038/s41467-020-20519-z
Keywords
-
Categories
Funding
- NSFC [61836004]
- Brain-Science Special Program of Beijing [Z181100001518006]
- Beijing Innovation Center for Future Chips
- Tsinghua University-China Electronics Technology HIK Group Co. Joint Research Center for Brain-inspired Computing, JCBIC
- National Science Foundation of China\National Natural Science Foundation of China-Yunnan Joint Fund (NSFC-Yunnan Joint Fund) [61327902]
- Nano Material Technology Development Program through the National Research Foundation of Korea (NRF) - Ministry of Science, ICT and Future Planning [2016M3A7B4910398]
- UNIST (Ulsan National Institute of Science and Technology) [1.200095.01]
- Artificial Intelligence Graduate School Program (UNIST) through the Institute of Information & communications Technology Planning & Evaluation (IITP) grant - Korea government (MSIT) [2020-0-01336]
- National Research Foundation of Korea [2016M3A7B4910398] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)
Ask authors/readers for more resources
The research team constructed a 39nm 1Gb phase-change memory memristor array and developed a spontaneous sparse learning scheme to improve PCM-based memristor network training. Experimental results show that this method helps enhance the performance and sparsity controllability of the network without requiring additional computation.
Neural networks trained by backpropagation have achieved tremendous successes on numerous intelligent tasks. However, naive gradient-based training and updating methods on memristors impede applications due to intrinsic material properties. Here, we built a 39nm 1Gb phase change memory (PCM) memristor array and quantified the unique resistance drift effect. On this basis, spontaneous sparse learning (SSL) scheme that leverages the resistance drift to improve PCM-based memristor network training is developed. During training, SSL regards the drift effect as spontaneous consistency-based distillation process that reinforces the array weights at the high-resistance state continuously unless the gradient-based method switches them to low resistance. Experiments show that the SSL not only helps the convergence of network with better performance and sparsity controllability without additional computation in handwritten digit classification. This work promotes the learning algorithms with the intrinsic properties of memristor devices, opening a new direction for development of neuromorphic computing chips.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available