On-device Continual Learning over Memory Hierarchy: A System Researcher’s Standpoint
Continual Learning (CL) is an emerging machine learning paradigm for edge devices that learn from a continuous stream of tasks. To avoid forgetting knowledge from previous tasks, episodic memory (EM) methods exploit a subset of the past samples while learning from new data. Despite promising results, prior studies are predominantly simulation-based and do not adequately address the growing demand for both EM capacity and system efficiency in practical setups. In this talk, I will discuss CarM, a system our research team has developed to meet this demand by employing hierarchical EM management as a key design principle. CarM uses high-speed RAMs for EM to ensure system efficiency and takes advantage of abundant storage to preserve past experiences, mitigating forgetting by facilitating efficient sample migration between memory and storage. Recently, we further improved CarM for cost-effectiveness, i.e., achieving high model accuracy without compromising energy efficiency, to make it more suitable for energy-sensitive edge devices. To gain insights into achieving cost-effectiveness, we first explore the design space of CarM. Miro, our new system runtime, carefully integrates these insights into CarM, enabling dynamic CL system configuration based on resource states for high cost-effectiveness. Miro also performs online profiling on parameters with clear accuracy-energy trade-offs and adapts to optimal values with minimal overhead. CL system research is still in its infancy, with numerous research problems to overcome to ensure deployability. I look forward to engaging in deeper and broader conversations during this seminar.
I am an Associate Professor in CSE at UNIST. In 2022 winter, I was a Visiting Professor in the DeepSpeed team at Microsoft. Prior to joining UNIST in 2018 fall, I spent several years in industry with Systems Research Group at Microsoft Research (2015-2018) and Systems Research Group at ARM Research (2014-2015). Currently at UNIST, I actively work on AI systems, real-time big data analytics, and systems for new HW, with my team members at OMNIA Lab. I finished my PhD in CS at Rice University, my MS in CS at KAIST, and my BE in CE at Kwangwoon University. I won the Best Paper Awards from ICDE 2022 (the first time in South Korea ever) and SYSTOR 2016.