EUVIMEDEUVIMED entdecken
Crossref

NeuroUNI: A Unified Event-Driven Multi-Core Architecture Optimizing Neuromorphic Primitives for Brain-Inspired Computing

Faquan Chen, Qingyang Tian, Ziren Wu, Xiangcheng Shi, Rendong Ying, Fei Wen, Peilin Liu

IEEE Transactions on Circuits and Systems I: Regular Papers · 2026-08 · 73 (8) · 5610-5623

DOI: 10.1109/tcsi.2026.3690710 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems I: Regular Papers, 2026-08. Verfasst von Faquan Chen und 6 weiteren Autor:innen. Die Arbeit nennt 59 Literaturangaben. Gefördert von: Brain Science and Brain-like Intelligence Technology—National Science and Technology Major Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The hardware convergence of Artificial Neural Networks (ANNs) and Spiking Neural Networks (SNNs) is hindered by conflicting computational paradigms: dense tensor parallelism versus asynchronous sparse dynamics. Existing unifications typically rely on inefficient spatial partitioning or mode-reconfigurable datapaths, limiting the flexibility needed by heterogeneous ANN-SNN hybrid models requiring frequent cross-domain interaction. To resolve this, we present NeuroUNI, a unified event-driven multi-core architecture. Unlike partitioned designs, NeuroUNI unifies computation at the primitive level using a novel Five-Tuple Event Model, abstracting both continuous activations and discrete spikes to naturally leverage dynamic sparsity. The architecture features a co-optimized hierarchical Macro-Micro- $\mu $ OP ISA, a superscalar SIMD-based microarchitecture, and a decentralized multi-core synchronization protocol. Validated in TSMC 28nm technology via post-synthesis simulation and on a Xilinx VCU129 FPGA prototype, NeuroUNI demonstrates competitive cross-paradigm efficiency. It achieves $35.0\times $ and $1.21\times $ the ANN energy efficiency of the NVIDIA V100 and EyerissV2, respectively, while delivering $5.7\times $ the SNN throughput of TrueNorth. In a unified mapless navigation workload, NeuroUNI attains 422.6 GOPS/W (ANN) and 190.8 GSOPS/W (SNN), outperforming Loihi 1 with $56.7\times $ the throughput and $3.85\times $ the energy efficiency, proving the viability of a primitive/ISA-level unified silicon substrate.

Förderung

  • Brain Science and Brain-like Intelligence Technology—National Science and Technology Major Project

Zitieren

APA

Chen, F., Tian, Q., Wu, Z., Shi, X., Ying, R., Wen, F., & Liu, P. (2026). NeuroUNI: A Unified Event-Driven Multi-Core Architecture Optimizing Neuromorphic Primitives for Brain-Inspired Computing. IEEE Transactions on Circuits and Systems I: Regular Papers, 73(8), 5610-5623. https://doi.org/10.1109/tcsi.2026.3690710

Vancouver

Chen F, Tian Q, Wu Z, Shi X, Ying R, Wen F, et al. NeuroUNI: A Unified Event-Driven Multi-Core Architecture Optimizing Neuromorphic Primitives for Brain-Inspired Computing. IEEE Transactions on Circuits and Systems I: Regular Papers. 2026;73(8):5610-5623. doi:10.1109/tcsi.2026.3690710

BibTeX
@article{Chen2026,
  author = {Chen, Faquan and Tian, Qingyang and Wu, Ziren and Shi, Xiangcheng and Ying, Rendong and Wen, Fei and Liu, Peilin},
  title = {NeuroUNI: A Unified Event-Driven Multi-Core Architecture Optimizing Neuromorphic Primitives for Brain-Inspired Computing},
  journal = {IEEE Transactions on Circuits and Systems I: Regular Papers},
  year = {2026},
  volume = {73},
  number = {8},
  pages = {5610--5623},
  doi = {10.1109/tcsi.2026.3690710}
}
Kostenlos starten →