利用 Miro 在内存层级上实现高性价比的端侧持续学习
Cost-effective On-device Continual Learning over Memory Hierarchy with Miro
AI总结：
本文提出 Miro 系统运行时，依据资源状态动态配置分层内存回放式持续学习，并在线剖析准确率-能耗权衡参数，从而在边缘设备上以低开销提升性价比。
摘要：
持续学习（CL）从连续的任务流中增量地训练神经网络（NN）模型。为了记住先前学到的知识，已有研究在内存层级上存储旧样本，并在新任务到达时回放这些样本。采用 CL 以保护数据隐私的边缘设备通常对能耗敏感，因此需要在不损害能效的前提下保持较高的模型准确率，即实现高性价比。我们的工作首次探索了基于分层内存回放的 CL 的设计空间，以获得在边缘设备上实现高性价比的洞见。我们提出了 Miro，这是一种新颖的系统运行时，通过使 CL 框架能够根据资源状态动态配置 CL 系统以获得最佳性价比，将我们的洞见谨慎地整合到 CL 框架中。为实现这一目标，Miro 还对具有明确准确率-能耗权衡的参数进行在线剖析，并以低开销适配到最优值。大量评估表明，Miro 显著优于我们为比较而构建的基线系统，持续实现更高的性价比。
英文摘要：
Continual learning (CL) trains NN models incrementally from a continuous stream of tasks. To remember previously learned knowledge, prior studies store old samples over a memory hierarchy and replay them when new tasks arrive. Edge devices that adopt CL to preserve data privacy are typically energy-sensitive and thus require high model accuracy while not compromising energy efficiency, i.e., cost-effectiveness. Our work is the first to explore the design space of hierarchical memory replay-based CL to gain insights into achieving cost-effectiveness on edge devices. We present Miro, a novel system runtime that carefully integrates our insights into the CL framework by enabling it to dynamically configure the CL system based on resource states for the best cost-effectiveness. To reach this goal, Miro also performs online profiling on parameters with clear accuracy-energy trade-offs and adapts to optimal values with low overhead. Extensive evaluations show that Miro significantly outperforms baseline systems we build for comparison, consistently achieving higher cost-effectiveness.
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。
