用图注意力网络+置信度评估，低成本高精度预测模型延迟，提升硬件感知神经架构搜索效率。
HiFi-LLP: High-Fidelity, Low-Cost Latency Predictors with Confidence for Robust HW-NAS
- 基于图注意力网络构建延迟预测模型，支持跨设备泛化。
- 在六个设备上达到0.996的秩相关系数，10%精度误差下比之前方法高9个百分点。
- 引入置信度机制，低置信预测自动转为真实硬件测试，提速8.6倍且性能不降。
随着深度神经网络（DNN）在边缘设备上的广泛应用，硬件感知优化技术（如硬件感知压缩和硬件感知神经架构搜索，HW-NAS）变得至关重要。这些方法依赖目标硬件的真实反馈来优化模型结构以实现高效部署。尽管搜索过程可并行化，但通过硬件在环（HIL）进行的延迟测量因串行性仍构成瓶颈。现有方法使用延迟预测器替代昂贵的HIL反馈，但仍面临挑战：(1) 平台特定的预测器通常需要数万个样本；(2) 预测不准可能误导NAS过程。为此，我们提出HiFi-LLP，一种基于图注意力网络的高保真、低成本延迟预测器，并引入置信度度量。在LatBench数据集的六个设备上，其最大斯皮尔曼秩相关系数达0.996，10%准确率边界下相比先前平台特定预测器提升高达9个百分点。我们进一步设计了一种混合NAS框架，将低置信度预测导向HIL，相比典型NAS实现最高8.6倍加速，同时保持竞争力的帕累托前沿。
原文摘要 · Abstract (English)
With deep neural networks (DNNs) increasingly deployed on edge devices, hardware (HW)-aware optimization techniques--such as HW-aware compression and HW-aware neural architecture search (HW-NAS)--have become essential. These methods rely on real feedback from the target hardware to tailor DNN architectures for efficient deployment. While the search can be parallelized, latency measurements via hardware-in-the-loop (HIL) remain a bottleneck due to their sequential nature. Recent approaches use latency predictors to replace costly HIL feedback, but challenges persist: (1) platform-specific predictors often require tens of thousands of samples, and (2) inaccurate predictions can mislead the NAS process. To address this, we introduce HiFi-LLP, a high-fidelity, low-cost latency predictor based on graph attention networks, augmented with a confidence metric. HiFi-LLP outperforms prior platform-specific predictors by up to 9 percentage points (p.p.) in the 10% accuracy bound and achieves a Spearman's rank correlation of up to 0.996 across six devices in the LatBench dataset. We further propose a hybrid NAS framework that routes low-confidence predictions to HIL, achieving up to 8.6$\times$ speedup compared to typical NAS while maintaining a competitive Pareto front.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。