English

Synapse Compression for Event-Based Convolutional-Neural-Network Accelerators

Hardware Architecture 2023-01-25 v3 Artificial Intelligence

Abstract

Manufacturing-viable neuromorphic chips require novel computer architectures to achieve the massively parallel and efficient information processing the brain supports so effortlessly. Emerging event-based architectures are making this dream a reality. However, the large memory requirements for synaptic connectivity are a showstopper for the execution of modern convolutional neural networks (CNNs) on massively parallel, event-based (spiking) architectures. This work overcomes this roadblock by contributing a lightweight hardware scheme to compress the synaptic memory requirements by several thousand times, enabling the execution of complex CNNs on a single chip of small form factor. A silicon implementation in a 12-nm technology shows that the technique increases the system's implementation cost by only 2%, despite achieving a total memory-footprint reduction of up to 374x compared to the best previously published technique.

Keywords

Cite

@article{arxiv.2112.07019,
  title  = {Synapse Compression for Event-Based Convolutional-Neural-Network Accelerators},
  author = {Lennart Bamberg and Arash Pourtaherian and Luc Waeijen and Anupam Chahar and Orlando Moreira},
  journal= {arXiv preprint arXiv:2112.07019},
  year   = {2023}
}

Comments

Preprint accepted by the IEEE Transactions on Parallel and Distributed Systems