arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.10658cs.ARcs.AIcs.LG
收藏

LUT-DLA:查找表作为高效的极低比特深度学习加速器

LUT-DLA: Lookup Table as Efficient Extreme Low-Bit Deep Learning Accelerator

Microsoft Research(微软研究院);University of Chinese Academy of Sciences(中国科学院大学);NTU Singapore(新加坡南洋理工大学)

Guoyu Li, Shengyu Ye, Chunyun Chen, Yang Wang, Fan Yang, Ting Cao, Cheng Liu, Mohamed M. Sabry, Mao Yang

AI总结:

本文提出LUT-DLA,一种基于查找表的深度学习加速器框架,通过向量量化实现极低比特量化,结合LUTBoost算法和协同设计优化,在保持较小精度损失下显著提升硬件效率。

摘要:

神经网络能力的出现,由于模型规模的扩大和计算复杂度的增加,不可避免地导致计算需求的显著激增。为了减小模型尺寸并降低推理成本,近期研究聚焦于简化模型和使用低比特量化设计硬件加速器。然而,由于数值表示的限制,标量量化无法将比特宽度降低到1比特以下,从而削弱了其优势。为了突破这些限制,我们提出了LUT-DLA,一种利用向量量化将神经网络模型转换为查找表(LUT)的深度学习加速器框架,实现了极低比特量化。LUT-DLA框架有助于设计高效且成本效益高的硬件加速器,并支持LUTBoost算法,该算法通过多阶段训练帮助将各种DNN模型转换为基于LUT的模型,大幅削减计算和硬件开销。此外,通过协同设计空间探索,LUT-DLA评估各种模型和硬件参数的影响,以针对不同应用场景微调硬件配置,优化性能和效率。我们的综合实验表明,LUT-DLA在功耗效率和面积效率上分别实现了1.4~7.0倍和1.5~146.1倍的提升,同时仅保持适度的精度下降。对于CNN,使用L2距离相似度时精度下降0.1%~3.1%,使用L1距离相似度时下降0.1%~3.4%,使用切比雪夫距离相似度时下降0.1%~3.8%。对于基于Transformer的模型,精度下降范围在1.4%到3.0%之间。

英文摘要:

The emergence of neural network capabilities invariably leads to a significant surge in computational demands due to expanding model sizes and increased computational complexity. To reduce model size and lower inference costs, recent research has focused on simplifying models and designing hardware accelerators using low-bit quantization. However, due to numerical representation limits, scalar quantization cannot reduce bit width lower than 1-bit, diminishing its benefits. To break through these limitations, we introduce LUT-DLA, a Look-Up Table (LUT) Deep Learning Accelerator Framework that utilizes vector quantization to convert neural network models into LUTs, achieving extreme low-bit quantization. The LUT-DLA framework facilitates efficient and cost-effective hardware accelerator designs and supports the LUTBoost algorithm, which helps to transform various DNN models into LUT-based models via multistage training, drastically cutting both computational and hardware overhead. Additionally, through co-design space exploration, LUT-DLA assesses the impact of various model and hardware parameters to fine-tune hardware configurations for different application scenarios, optimizing performance and efficiency. Our comprehensive experiments show that LUT-DLA achieves improvements in power efficiency and area efficiency with gains of $1.4$~$7.0\times$ and $1.5$~$146.1\times$, respectively, while maintaining only a modest accuracy drop. For CNNs, accuracy decreases by $0.1\%$~$3.1\%$ using the $L_2$ distance similarity, $0.1\%$~$3.4\%$ with the $L_1$ distance similarity, and $0.1\%$~$3.8\%$ when employing the Chebyshev distance similarity. For transformer-based models, the accuracy drop ranges from $1.4\%$ to $3.0\%$.

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。

↑