Advancing reliable
and responsible
AI systems
Sun Research Lab investigates bias and uncertainty dynamics in multi-stage learning pipelines, adaptive AI workload management, and responsible AI evaluation — building a foundation for fairer, more dependable machine learning.
Two research tracks
AI Bias & Fairness
Investigating how bias originates, propagates, and compounds through machine learning pipelines — and building frameworks to detect, audit, and correct it.
Bias and uncertainty in multi-stage learning pipelines
Identifying amplification and identifiability regimes across ML pipeline stages.
Evaluation frameworks for responsible AI systems
Principled methodologies for assessing fairness, reliability, and societal impact.
Software reliability in AI and data-processing systems
Failure modes, testability, and automated test generation for AI software.
AI Systems & IoT
Designing adaptive, energy-efficient scheduling and resource management strategies for AI systems operating under real-time and resource-constrained conditions.
Adaptive resource management for variable AI workloads
Scheduling and allocation policies that respond to unpredictable workload fluctuations.
Power-efficient scheduling for IoT and edge AI
DVFS and NVRAM strategies for real-time energy-efficient task scheduling.
Tight schedulability analysis for hybrid memory systems
Formal analysis combining processor and memory power techniques.
Research output
Published
Adaptive Swapping for Variable Workloads in Real-Time Task Scheduling
Nam, S. A. & Bahn, H. — IEEE CCCI 2023, Chongqing, China
A Resource Planning Policy to Support Variable Real-Time Tasks in IoT Systems
Bahn, H. & Nam, S. A. — Journal of Internet, Broadcasting and Communication, 23(4)
Adaptive Resource Planning for AI Workloads with Variable Real-Time Tasks
Nam, S. A., Cho, K., & Bahn, H. — Computers, Materials & Continua, 74(3), 6823–6833
A New Resource Configuring Scheme for Variable Workloads in IoT Systems
Nam, S. A., Cho, K., & Bahn, H. — IEEE CSDE 2022, Queensland, Australia
Tight Evaluation of Real-Time Task Schedulability for Processor's DVS and Nonvolatile Memory Allocation
Nam, S. A., Cho, K., & Bahn, H. — Micromachines, 10(6), 371–382
Combining Memory Allocation and Processor Voltage Scaling for Energy Efficient IoT Scheduling
Nam, S. A., Cho, K., & Bahn, H. — IEEE/ACIS ICIS 2017, Wuhan, China
Real-Time Task Scheduling Methods to Incorporate Power Techniques of Processors and Memory in IoT Environments
Nam, S. A. & Bahn, H. — Journal of Internet, Broadcasting and Communication, 17(2), 1–6
Researchers
Dr. Sunhwa A. Nam
Ph.D. in Computer Science, Ewha Womans University. Research spans AI systems engineering, bias dynamics in learning pipelines, and adaptive resource management. Former Research Professor at Ewha Womans University and Patent Examiner at the USPTO. IEEE Member.
Dr. Jihyun Park
Collaborating researcher in machine learning fairness and bias propagation, with joint research in stage-aware bias dynamics in ML pipelines.
Get in touch
Collaborate, connect, or learn more
We welcome collaboration with academic institutions, industry partners, and policy organisations working on responsible AI, bias research, and AI systems engineering.