低功耗设备端持续学习加速器，能效比超当前方案7倍以上。
Clo-HDnn: A 4.66 TFLOPS/W and 3.78 TOPS/W Continual On-Device Learning Accelerator with Energy-efficient Hyperdimensional Computing via Progressive Search
- 用高维计算与渐进搜索优化模型更新，免于梯度计算。
- 在4.66 TFLOPS/W和3.78 TOPS/W下实现7.77倍能效提升。
- 适合资源受限的边缘设备持续学习任务。
Clo-HDnn是一种面向新兴持续学习任务的设备端学习加速器。通过融合高维计算（HDC）、低成本克罗内克编码器与权重聚类特征提取（WCFE），在保证精度的同时优化效率。采用无梯度持续学习机制，以类别超向量形式高效存储知识。双模式运行可跳过简单数据集的复杂特征提取，渐进搜索则通过仅编码和比较部分查询超向量，将计算复杂度降低高达61%。在分类器模式下达到3.78 TOPS/W，前向执行模式达4.66 TFLOPS/W，能效分别较当前最优方案提升7.77倍和4.85倍。
原文摘要 · Abstract (English)
Clo-HDnn is an on-device learning (ODL) accelerator designed for emerging continual learning (CL) tasks. Clo-HDnn integrates hyperdimensional computing (HDC) along with low-cost Kronecker HD Encoder and weight clustering feature extraction (WCFE) to optimize accuracy and efficiency. Clo-HDnn adopts gradient-free CL to efficiently update and store the learned knowledge in the form of class hypervectors. Its dual-mode operation enables bypassing costly feature extraction for simpler datasets, while progressive search reduces complexity by up to 61% by encoding and comparing only partial query hypervectors. Achieving 4.66 TFLOPS/W (FE) and 3.78 TOPS/W (classifier), Clo-HDnn delivers 7.77x and 4.85x higher energy efficiency compared to SOTA ODL accelerators.
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。