arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.01546cs.AR
收藏

神经AI时序神经网络(NeuTNNs):面向专用神经形态处理单元的微架构与设计框架

NeuroAI Temporal Neural Networks (NeuTNNs): Microarchitecture and Design Framework for Specialized Neuromorphic Processing Units

Carnegie Mellon University(卡内基梅隆大学)

Shanmuga Venkatachalam, Prabhu Vellaisamy, Harideep Nair, Wei-Che Huang, Youngseok Na, Yuyang Kang, Quinn Jacobson, John Paul Shen

AI总结:

本文提出神经AI时序神经网络(NeuTNNs)及设计工具NeuTNNGen,通过神经科学发现提升性能与硬件效率,适用于多种应用场景。

摘要:

来自两个领域的领先专家已建议需要重新连接神经科学和人工智能(AI)的研究,以加速下一代AI创新的发展。他们将这种融合称为神经AI。先前的研究已经确立了时序神经网络(TNNs)作为一种有前途的神经形态方法,以实现生物智能和效率。我们完全拥抱神经AI,并提出了一种新的TNNs类别,称为神经AI TNNs(NeuTNNs),通过采用神经科学发现,包括具有主动树突的神经元模型和远端和近端段的层次结构,从而实现更大的能力和硬件效率。这项工作介绍了一套从PyTorch到布局的工具套件(NeuTNNGen),用于设计应用特定的NeuTNNs。与先前的TNN设计相比,NeuTNNs在性能和效率上均优于前者。我们通过三个示例应用展示了NeuTNNGen的能力:1)UCR时间序列基准测试,2)MNIST设计探索,3)皮层参考框架的Place Cells设计。我们还探讨了使用突触剪枝进一步减少突触计数和硬件成本30-50%,同时在多样化的感官模态中保持模型精度。NeuTNNGen可以促进下一代神经AI计算系统应用特定的节能NeuTNNs的设计。

英文摘要:

Leading experts from both communities have suggested the need to (re)connect research in neuroscience and artificial intelligence (AI) to accelerate the development of next-generation AI innovations. They term this convergence as NeuroAI. Previous research has established temporal neural networks (TNNs) as a promising neuromorphic approach toward biological intelligence and efficiency. We fully embrace NeuroAI and propose a new category of TNNs we call NeuroAI TNNs (NeuTNNs) with greater capability and hardware efficiency by adopting neuroscience findings, including a neuron model with active dendrites and a hierarchy of distal and proximal segments. This work introduces a PyTorch-to-layout tool suite (NeuTNNGen) to design application-specific NeuTNNs. Compared to previous TNN designs, NeuTNNs achieve superior performance and efficiency. We demonstrate NeuTNNGen's capabilities using three example applications: 1) UCR time series benchmarks, 2) MNIST design exploration, and 3) Place Cells design for neocortical reference frames. We also explore using synaptic pruning to further reduce synapse counts and hardware costs by 30-50% while maintaining model precision across diverse sensory modalities. NeuTNNGen can facilitate the design of application-specific energy-efficient NeuTNNs for the next generation of NeuroAI computing systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑