LLM指导的框架优化物理神经网络架构

作者 PulseAugur 编辑部 · [2 个来源] · 2026-06-09 01:32

研究人员开发了一个名为UH-NAS的新框架，该框架使用LLM来指导物理神经网络的神经架构搜索。这种方法将任务准确性与能耗和物理非理想性等硬件约束进行了协同优化。UH-NAS被设计为硬件无关的，能够跨不同计算平台进行公平比较，并发现比传统方法更鲁棒的架构。 AI

排序理由该集群包含一篇详细介绍神经架构搜索新方法的学术论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.AI TIER_1 English(EN) · Tyler King, Timothee Leleu · 2026-06-10 04:00

LLM-Guided Neural Architecture Search for Robust Co-Design of Physical Neural Networks

arXiv:2606.10294v1 Announce Type: cross Abstract: Deploying neural networks on unconventional hardware demands architectures that co-optimize task accuracy and platform-specific constraints such as energy cost, physical non-idealities, and numerical precision. Existing neural arc…
arXiv cs.NE (Neural & Evolutionary) TIER_1 English(EN) · Timothee Leleu · 2026-06-09 01:32

LLM 引导的神经架构搜索用于物理神经网络的鲁棒协同设计

Deploying neural networks on unconventional hardware demands architectures that co-optimize task accuracy and platform-specific constraints such as energy cost, physical non-idealities, and numerical precision. Existing neural architecture search (NAS) methods are typically tailo…