Research
SLED Lab develops system software for computing platforms that interact with the physical world. We combine learning, scheduling, and domain knowledge to make edge, battery-powered, and AI systems more adaptive, efficient, and reliable.
A. Learning-based system optimization
Keywords: reinforcement learning · on-device AI · healthcare
We develop learning-based systems that adapt to changing workloads and environments. Our work spans resource management and on-device applications, including power control, energy prediction, and contactless health monitoring. These methods support adaptive operation across mobile and embedded platforms.
Representative:
EarDVFS: Environment-Adaptable RL-based DVFS… (ICCAD 2025)
mCardiacDx: Radar-Driven Contactless Monitoring… (JTEHM 2026)
Serenus: Alleviating Low-Battery Anxiety… (UIST 2024)
B. System-level support for battery & energy
Keywords: battery management systems · embedded systems · energy efficiency
We design system-level methods that improve how batteries are configured, charged, and used across diverse applications. Our work supports efficient, reliable operation in mobile devices, large-scale battery systems, and electric-vehicle infrastructure.
Representative:
MixMax: Leveraging Heterogeneous Batteries… (MobiSys 2023)
Leveraging Customized Heterogeneous Batteries… (TSUSC 2025)
Scheduling EV Battery Swap/Charge Operations (RTAS 2025)
RAC+: Supporting Reconfiguration-Assisted Charging… (TII 2024)
C. Novel resource management framework
Keywords: scheduling algorithms · AI agents · LLM serving
We develop scheduling and resource-management frameworks that provide predictable performance while improving energy efficiency and system lifetime. We are extending these ideas to emerging AI-agent and LLM-serving workloads.
Representative:
Battery Aging Deceleration… (RTSS 2019)
Non-Preemptive Real-Time Multiprocessor Scheduling… (RTSS 2020)
Battery-Aging-Aware Run-Time Slack Management… (JSA 2023) CoRT: Supporting Hard and Soft Real-Time Tasks… (ICCAD 2026)
