Deep neural network (DNN) video analytics is crucial for autonomous systems such as self-driving vehicles, unmanned aerial vehicles (UAVs), and security robots. However, real-world deployment faces challenges due to their limited computational resources and battery power. To tackle these challenges, continuous learning exploits a lightweight "student" model at deployment (inference), leverages a larger "teacher" model for labeling sampled data (labeling), and continuously retrains the student model to adapt to changing scenarios (retraining). This paper highlights the limitations in state-of-the-art continuous learning systems: (1) they focus on computations for retraining, while overlooking the compute needs for inference and labeling, (2) they rely on power-hungry GPUs, unsuitable for battery-operated autonomous systems, and (3) they are located on a remote centralized server, intended for multi-tenant scenarios, again unsuitable for autonomous systems due to privacy, network availability, and latency concerns. We propose a hardware-algorithm co-designed solution for continuous learning, DaCapo, that enables autonomous systems to perform concurrent executions of inference, labeling, and training in a performant and energy-efficient manner. DaCapo comprises (1) a spatially-partitionable and precision-flexible accelerator enabling parallel execution of kernels on sub-accelerators at their respective precisions, and (2) a spatiotemporal resource allocation algorithm that strategically navigates the resource-accuracy tradeoff space, facilitating optimal decisions for resource allocation to achieve maximal accuracy. Our evaluation shows that DaCapo achieves 6.5% and 5.5% higher accuracy than a state-of-the-art GPU-based continuous learning systems, Ekya and EOMU, respectively, while consuming 254x less power.

翻译：深度神经网络（DNN）视频分析对于自动驾驶车辆、无人机（UAV）和安全机器人等自主系统至关重要。然而，实际部署面临计算资源和电池容量有限的挑战。为此，持续学习技术在部署阶段（推理）采用轻量级"学生"模型，借助更大型的"教师"模型为采样数据标注（标注），并通过持续重新训练学生模型以适应场景变化（再训练）。本文指出现有持续学习系统的局限：（1）它们专注于再训练计算，却忽视推理和标注的计算需求；（2）依赖高功耗GPU，不适用于电池供电的自主系统；（3）部署在远程集中式服务器上，适用于多租户场景，却因隐私、网络可用性和延迟问题同样不适用自主系统。我们提出硬件-算法协同设计的持续学习方案DaCapo，使自主系统能够以高性能和高能效方式并行执行推理、标注和训练。DaCapo包含：（1）空间可分区、精度可调节的加速器，支持在子加速器上以各自精度并行执行内核；（2）时空资源分配算法，通过策略性权衡资源-精度空间，做出最优资源分配决策以实现最高精度。评估表明，相比基于GPU的先进持续学习系统Ekya和EOMU，DaCapo在功耗降低254倍的同时，分别实现6.5%和5.5%的精度提升。