Modern systems-on-chip (SoCs) use dynamic power management (DPM) techniques to improve energy efficiency. However, existing techniques are unable to efficiently adapt the runtime decisions considering multiple objectives (e.g., energy and real-time requirements) simultaneously on heterogeneous platforms. To address this need, we propose HiLITE, a hierarchical imitation learning framework that maximizes the energy efficiency while satisfying soft real-time constraints on embedded SoCs. Our approach first trains DPM policies using imitation learning; then, it applies a regression policy at runtime to minimize deadline misses. HiLITE improves the energy-delay product by 40 percent on average, and reduces deadline misses by up to 76 percent, compared to state-of-the-art approaches. In addition, we show that the trained policies not only achieve high accuracy, but also have negligible prediction time overhead and small memory footprint.

中文翻译：

---

HiLITE：用于嵌入式 SoC 电源管理的分层轻量级模仿学习

现代片上系统 (SoC) 使用动态电源管理 (DPM) 技术来提高能效。然而，现有技术无法在异构平台上同时考虑多个目标（例如，能量和实时要求）有效地调整运行时决策。为了满足这一需求，我们提出了 HiLITE，这是一种分层模仿学习框架，可最大限度地提高能源效率，同时满足嵌入式 SoC 的软实时约束。我们的方法首先使用模仿学习来训练 DPM 策略；然后，它在运行时应用回归策略以最小化截止期限未命中。与最先进的方法相比，HiLITE 将能量延迟产品平均提高了 40%，并将最后期限未命中率降低了 76%。此外，