Data-Driven Neuromorphic DRAM-based CNN and RNN Accelerators

Tobi Delbruck; Shih-Chii Liu

Data-Driven Neuromorphic DRAM-based CNN and RNN Accelerators

Computer Vision and Pattern Recognition 2020-03-31 v1 Neural and Evolutionary Computing

Authors: Tobi Delbruck , Shih-Chii Liu

Abstract

The energy consumed by running large deep neural networks (DNNs) on hardware accelerators is dominated by the need for lots of fast memory to store both states and weights. This large required memory is currently only economically viable through DRAM. Although DRAM is high-throughput and low-cost memory (costing 20X less than SRAM), its long random access latency is bad for the unpredictable access patterns in spiking neural networks (SNNs). In addition, accessing data from DRAM costs orders of magnitude more energy than doing arithmetic with that data. SNNs are energy-efficient if local memory is available and few spikes are generated. This paper reports on our developments over the last 5 years of convolutional and recurrent deep neural network hardware accelerators that exploit either spatial or temporal sparsity similar to SNNs but achieve SOA throughput, power efficiency and latency even with the use of DRAM for the required storage of the weights and states of large DNNs.

Keywords

deep neural network acceleration spiking neural network neuromorphic computing

Cite

@article{arxiv.2003.13006,
  title  = {Data-Driven Neuromorphic DRAM-based CNN and RNN Accelerators},
  author = {Tobi Delbruck and Shih-Chii Liu},
  journal= {arXiv preprint arXiv:2003.13006},
  year   = {2020}
}

Comments

To appear in 2019 IEEE Sig. Proc. Soc. Asilomar Conference on Signals, Systems, and Computers Session MP6b: Neuromorphic Computing (Invited)

Data-Driven Neuromorphic DRAM-based CNN and RNN Accelerators

Abstract

Keywords

Cite

Comments

Related papers