English

A Diagonal Structured State Space Model on Loihi 2 for Efficient Streaming Sequence Processing

Machine Learning 2024-09-24 v1 Artificial Intelligence Emerging Technologies Neural and Evolutionary Computing

Abstract

Deep State-Space Models (SSM) demonstrate state-of-the art performance on long-range sequence modeling tasks. While the recurrent structure of SSMs can be efficiently implemented as a convolution or as a parallel scan during training, recurrent token-by-token processing cannot currently be implemented efficiently on GPUs. Here, we demonstrate efficient token-by-token inference of the SSM S4D on Intel's Loihi 2 state-of-the-art neuromorphic processor. We compare this first ever neuromorphic-hardware implementation of an SSM on sMNIST, psMNIST, and sCIFAR to a recurrent and a convolutional implementation of S4D on Jetson Orin Nano (Jetson). While we find Jetson to perform better in an offline sample-by-sample based batched processing mode, Loihi 2 outperforms during token-by-token based processing, where it consumes 1000 times less energy with a 75 times lower latency and a 75 times higher throughput compared to the recurrent implementation of S4D on Jetson. This opens up new avenues towards efficient real-time streaming applications of SSMs.

Keywords

Cite

@article{arxiv.2409.15022,
  title  = {A Diagonal Structured State Space Model on Loihi 2 for Efficient Streaming Sequence Processing},
  author = {Svea Marie Meyer and Philipp Weidel and Philipp Plank and Leobardo Campos-Macias and Sumit Bam Shrestha and Philipp Stratmann and Mathis Richter},
  journal= {arXiv preprint arXiv:2409.15022},
  year   = {2024}
}

Comments

6 pages, 2 figures

R2 v1 2026-06-28T18:53:43.911Z