Skip to content
Book Open access

Temporal Sparse Die-to-Die Communication in Heterogeneous Neuromorphic Systems

Aug 2026 · International Conference on Systems · 0 citations · 37 references
Computer Science

TL;DR

This work proposes heterogeneous neural networks that combine spiking neural networks (SNNs) and artificial neural networks (ANNs) at bandwidth-limited regions, such as chip boundaries, where spike-based communication reduces data transfer overhead.

Abstract

Efficient communication is central to both biological and artificial intelligence (AI) systems. In biological brains, the challenge of long-range communication across regions is addressed through sparse, spike-based signaling, minimizing energy and latency. Conversely, modern AI workloads are increasingly constrained by bandwidth, leading to bottlenecks that hamper scalability and efficiency. Inspired by the brain’s ability to execute dynamic and complex local computations coupled with sparse inter-neuron communication, we propose heterogeneous neural networks that combine spiking neural networks (SNNs) and artificial neural networks (ANNs) at bandwidth-limited regions, such as chip boundaries, where spike-based communication reduces data transfer overhead. Within each chip, dense ANN computations maintain high throughput, accuracy, and robustness. While SNNs have struggled to algorithmically scale, our approach surmounts this long-standing challenge through algorithm-architecture co-design where learnable sparsity is employed for die-to-die communication by confining spiking layers to specific partitions. This composable design combines high ANN performance with low-bandwidth SNN efficiency. Evaluations on language processing and computer vision exhibit up to 5.3 × energy efficiency gains and 15.2 × latency reductions, surpassing both purely spiking and non-spiking models. As model size grows, improvements scale accordingly. By targeting the inter-chip communication bottleneck with biologically inspired methods, this approach presents a promising path to more efficient AI systems.

Read PDF

Similar papers

Preprint Aug 2026

Neuromorphic architectures as numerical solvers for computational neuroscience

Neuromorphic computing is closely associated with spiking neuronal networks. However, an alternative class of so-called"rate-based"models arising from computational neuroscience and machine learning forgoes spiking interactions and instead relies on continuous coupling between neurons. Existing neuromorphic implementations designed around spike-based interactions are not well-suited for emulating such models. Here view the distributed simulation of these models as message-passing algorithms on parallel hardware. Leveraging prior art in numerical algorithms and distributed simulation, we outline steps that enable the design of efficient digital neuromorphic accelerators for non-spiking neuronal models. In particular, we show that multi-bit packets, rather than spikes, are the most efficient communication strategy in packet-switched networks and that compared to basic numerical integration methods, higher-order differential equation solvers decrease both computation and communication costs while achieving lower numerical error, but that these benefits are ultimately limited by arithmetic precision. Using our proposed design principles, we convert an existing neuromorphic architecture into a distributed numerical solver - a spikeless neuromorphic system - for continuously-coupled neuronal models. We thereby demonstrate that our theoretical considerations indeed translate into practical advantages, namely reduced energy consumption and delay.

Jakob Jordan, Ole Richter, Cong Li et al. · 0 citations
Preprint Aug 2026

Small-World Communication Fabrics for Neuromorphic Multicore-SoCs

As neuromorphic systems scale beyond a single core, inter-core event communication can become a dominant contributor to memory footprint, latency, and energy consumption. Biological neural systems address a similar scaling challenge through small-world organization, combining dense local connectivity with sparse long-range projections. In this work, we compare two recent multicore neuromorphic systems implemented in the same 22-nm FDSOI technology and explicitly optimized for such connectivity. The first, NeoCorAl, uses an asynchronous packet-switched tree with hierarchical multicast, whereas the second, MOSAIC, employs an RRAM-based, circuit-switched two-dimensional mesh that performs routing in memory. We examine the resulting trade-offs in routing flexibility, hop count, memory requirements, multicast efficiency, and scalability. We further study how the relative efficiency of tree- and mesh-based routing depends on communication locality in spatially-embedded, random, and layered networks. Finally, we discuss routing-aware training as a means of jointly optimizing neural connectivity, task performance, and hardware mappability.

Sebastian Billaudelle, Christian Metzner, Jimmy Weber et al. · 0 citations
Review Jul 2026

Spiking Neural Networks: A Computational Paradigm for Neuromorphic Computing

It is aimed at proving that SNNs have potential in such areas as computer vision, robotics, and speech recognition, and their role in overcoming the barrier between artificial and biological neural systems is proved.

Mesala Sravani, K. Kumari, S. M. Reddy · 0 citations
Open access Aug 2026

Neuromorphic RISC-V Systems for Bio-inspiredComputing Applications

Neuromorphic computing is a paradigm based on the computational mechanisms of the human brain and hasreceived considerable attention as a real-time technique with low energy requirements. Present systems, however, arelimited in their ability to scale traditional processors to a neuromorphic architecture, leading to issues with latency,power consumption, and smooth data flow. To address these problems, this paper proposes the ACORISC-VbSNNframework, comprising a modular RISC-V architecture, Spiking Neural Networks (SNNs), and Ant ColonyOptimization (ACO). The system uses a shared-memory architecture to maximize communication between traditionaland neuromorphic processors, ensuring data is managed effectively. The postulated framework processes the sensorydata by pre-processing and encoding them using rate coding, and dynamically optimizing memory access. SNNs arealso used to process spike trains in real-time, whereas ACO is used to determine the best data paths to minimizebottlenecks. Experimental analysis shows that the system performs better, with ultra-low power consumption of 0.0095mW, very low latency of 0.000544 seconds, and 99.2 percent accuracy. These findings indicate that the ACORISCVbSNN model has the potential to advance the field of bio-inspired computing, providing a highly accurate, energyefficient, and low-latency system for real-world use.

Yamini Devi Ykuntam, M. V. Nageswara Rao, Leela Kumari. B. · 0 citations
Preprint Aug 2026

BASC : Behavior-Aligned Quantization and Pruning for Low-Bit Spiking Neural Networks

Extensive experiments on static and neuromorphic benchmarks show that lower-bit BASC models match or outperform higher-bit baselines and retain this accuracy advantage after structured pruning, while further reducing model storage and synaptic operations.

Linliang Chen, Yan Zhong, Xin Liu et al. · 0 citations
Open access Jul 2026

Community-aware sparse topology design for efficient spiking neural networks.

This work proposes a community-aware sparse topology design framework for graph-based SNNs and challenges the prevailing assumption that network size or sparsity alone is sufficient, demonstrating that how sparse connections are organized - the graph topology - critically influences learning efficiency, accuracy, and energy consumption.

F. Motaghian, S. Nazari, J. P. Dominguez-Morales et al. · 0 citations