Pyeongjun Choi

• Ph.D. Student, Department of Electrical Engineering & Computer Science, DGIST. Advisor: Jeongho Kwak. Mar. 2020 – Present.
• B.S., School of Undergraduate Studies, DGIST. Mar. 2016 – Feb. 2020.

• Resource allocation in mobile edge computing system
• Mobile-cloud collaborative inference
• Thermal- and memory-aware mobile AI

• Bronze Prize, Samsung Humantech Paper Award, Feb. 2022.
• Bronze Prize, Samsung Humantech Paper Award, Feb. 2021.
• JKICS Best Paper Award (March 2024 issue).
• LG U+ Industry–Academia Scholarship, Nov. 2024.
• Graduate School Outstanding Student Award.
• DGIST Student Conference Outstanding Poster.
• Korea Government Full Scholarship (8 consecutive undergraduate semesters).

Journal Articles (SCIE)

1. Pyeongjun Choi^†, Dongho Ham^†, Yeongjin Kim, and Jeongho Kwak, “VisionScaling: Dynamic Deep Learning Model and Resource Scaling in Mobile Vision Applications,” IEEE Internet of Things Journal, vol. 11, no. 9, pp. 15523–15539, May 2024. (IF: 10.6, JCR Top 3% in Computer Science, Information Systems).
2. Yeongjin Kim, Pyeongjun Choi, Jeong-A Lim, and Jeongho Kwak, “Network-Compute Co-Optimization for Service Chaining in Cloud-Edge-Radio 5G Networks,” IEEE Transactions on Vehicular Technology, vol. 72, no. 10, pp. 13374–13391, Oct. 2023. (IF: 7.5, JCR Top 13% in Telecommunications).
3. Sinuk Choi, Pyeongjun Choi, Donghyeon Kim, Jeongho Kwak, and Ji-Woong Choi, “An Integrated Process-Network Load Balancing in Edge-Assisted Autonomous Vehicles Using Multimodal Applications With Shared Workloads,” IEEE Access, vol. 12, pp. 174654–174667, Nov. 2024.
4. “DuraMAC: A Thermally Durable AI Inference Framework via Mobile and Cloud Co-Execution,” in preparation.

Domestic Journals

1. Pyeongjun Choi, Pildo Yoon, and Jeongho Kwak, “Dynamic Load Balancing Algorithm for Energy-Delay Tradeoff in a Cloud-RSU-Vehicle Architecture,” Journal of Korean Institute of Communications and Information Sciences (JKICS), vol. 49, no. 3, pp. 377–384, Mar. 2024. (Best Paper Award).

International Conferences

1. Omitted due to double blind review, “AutoDraft: Automatic Cost-Performance Adaptation in User-Cloud Distributed Speculative Decoding,” submitted to conference.
2. Pyeongjun Choi, Jeongsoo Kim, Seyeon Kim, and Jeongho Kwak, “DualEngine: A Thermal-Aware Vision Inference Framework via Mobile and Cloud Co-Execution,” in Proc. of IEEE SECON, Pisa, Italy, May 2026, pp. 70–78. (BK21+ Outstanding International Conference).
3. Eunsu Kim, Pyeongjun Choi, Jeonghwan Kim, and Jeongho Kwak, “Interchangeable CPU-GPU DVFS for Encoding vs. Inference in LEO Satellite Onboard Processing,” in Proc. of IEEE/IFIP NOMS, Rome, Italy, May 2026, pp. 1–10. (BK21+ Outstanding International Conference).
4. Pyeongjun Choi, Jeongsoo Kim, and Jeongho Kwak, “Joint Task Offloading and Resource Allocation for Integrated V2V and V2I Communication,” in Proc. of ICTC, Jeju, Korea, Oct. 2024.
5. Pyeongjun Choi, Jeongsoo Kim, and Jeongho Kwak, “Impact of Joint Heat and Memory Constraints of Mobile Device in Edge-Assisted On-Device Artificial Intelligence,” in Proc. of the 2nd International Workshop on Networked AI Systems (NetAISys), Tokyo, Japan, Jun. 2024.
6. Pyeongjun Choi, Pildo Yoon, and Jeongho Kwak, “Dynamic Load Balancing for Energy-Delay Tradeoff in a Cloud-RSU-Vehicle Architecture,” in Proc. of ICTC, Jeju, Korea, Oct. 2023.
7. Pyeongjun Choi and Jeongho Kwak, “A Survey on Mobile Edge Computing for Deep Learning,” in Proc. of ICOIN, Bangkok, Thailand, Jan. 2023.

Domestic Conferences

1. Eunsu Kim, Pyeongjun Choi, Jeonghwan Kim, and Jeongho Kwak, “CPU-GPU DVFS-based Image Inference and Code Offloading for LEO Satellite Onboard Systems,” KICS Fall Conference, Gyeongju, Korea, Nov. 2025.
2. Jeongsoo Kim, Pyeongjun Choi, and Jeongho Kwak, “Impact of Thermal and Memory Constraints of Mobile Devices in Edge-Assisted On-Device AI,” KICS Summer Conference, Jeju, Korea, Jun. 2024.
3. Pyeongjun Choi, Jeongsoo Kim, and Jeongho Kwak, “Impact of Heat and Memory on Deep Learning Application Performance in Mobile AI,” KICS Winter Conference, Yongpyeong, Korea, Jan. 2024.
4. Pyeongjun Choi and Jeongho Kwak, “A Study on Stable Mobile AI,” KICS Summer Conference, Jeju, Korea, Jun. 2023.
5. Pyeongjun Choi, Pildo Yoon, and Jeongho Kwak, “Dynamic Computing Load Balancing Algorithm between Vehicles and RSUs,” JCCI, Yeosu, Korea, Apr. 2023.
6. Pyeongjun Choi, Jeongmin Seo, and Jeongho Kwak, “Dynamic Deep Learning Model and Resource Scaling for Multi-User Mobile Vision Applications,” KICS Winter Conference, Yongpyeong, Korea, Feb. 2023.
7. Pyeongjun Choi, Dongho Ham, Yeongjin Kim, and Jeongho Kwak, “A Study on the Impact of Joint Learning Model and Resource Control in Mobile AI,” KICS Winter Conference, Pyeongchang, Korea, 2022.
8. Pyeongjun Choi, Jinhwi Kim, Yeongjin Kim, and Jeongho Kwak, “Efficiency Analysis of CPU-GPU Scaling in Embedded AI,” JCCI, Busan, Korea, Apr. 2021.

Patents (Overseas)

1. Jeongho Kwak, Ji-Woong Choi, Pyeongjun Choi, Sinuk Choi, and Pildo Yoon, “Method and Device for Determining Optimal Load Balancing in a Vehicle,” PCT/KR2023/021288.
2. Jeongho Kwak, Ji-Woong Choi, Pyeongjun Choi, Sinuk Choi, and Pildo Yoon, “METHOD AND DEVICE FOR DETERMINING OPTIMAL LOAD BALANCING IN A VEHICLE,” US 18/856,960.
3. Jeongho Kwak, Ji-Woong Choi, Pyeongjun Choi, Sinuk Choi, and Pildo Yoon, “METHOD AND DEVICE FOR DETERMINING OPTIMAL LOAD BALANCING IN A VEHICLE,” EP 23931188.9.
4. Jeongho Kwak, Ji-Woong Choi, Pyeongjun Choi, Sinuk Choi, and Pildo Yoon, “車両内最適なロードバランシング決定方法及び装置,” JP 2024-560940.

Patents (Domestic)

1. Jeongho Kwak, Eunsu Kim, Jeonghwan Kim, and Pyeongjun Choi, “위성 온보드 시스템에서 적응적 인코딩-추론 및 CPU/GPU DVFS 제어 방법, 이를 수행하는 시스템 및 컴퓨터 프로그램,” KR 10-2026-0143545.
2. Pyeongjun Choi, Jeongho Kwak, Dongho Ham, and Yeongjin Kim, “Method and Apparatus for Simultaneously Optimizing Terminal Resources and Learning Models,” KR 10-2023-0044201.
3. Jeongho Kwak, Ji-Woong Choi, Pyeongjun Choi, Sinuk Choi, and Pildo Yoon, “Method and Apparatus for Determining Optimal In-Vehicle Load Balancing,” KR 10-2023-0142928.
4. Jeongho Kwak, Pyeongjun Choi, and Jeongsoo Kim, “Method and Apparatus for Processing Tasks Considering Resource Constraints of User Terminals,” KR 10-2024-0179020.
5. “사용자 단말-클라우드 공동 실행 기반 서비스 수준 목표 인식형 적응형 분산 투기 디코딩 방법 및 시스템,” in preparation.

Software Registration

• Korea University Industry–University Cooperation Foundation (Inventor share: Jeongho Kwak 50%, Pyeongjun Choi 50%), “Vision AI Program for Thermal Stabilization Using Mobile–Cloud Collaborative Inference.”

• VisionScaling: 6G 비전 서비스를 위한 모바일에서 클라우드까지 확장 가능한 통합 학습모델 및 자원 스케일링 프레임워크 개발용 클라우드 서버 구축, 최초혁신실험실 추가지원, 한국연구재단 우수신진연구, 2022.03 – 2027.02.
• 다중 통신기술 네트워크 로드밸런싱 기술 개발, 정보통신기술진흥센터 (IITP), Researcher, 2022.04 – 2024.12.

• Journal reviewer: IEEE Transactions on Wireless Communications; IEEE Transactions on Vehicular Technology; IEEE/ACM Transactions on Networking.
• Conference reviewer: IEEE WiOpt 2023; IEEE Globecom 2024, 2025.

• Android programming; Python; MATLAB
• On-device / mobile AI systems (thermal-, memory-aware inference, DVFS)
• LLM system implementation; mobile–cloud collaborative inference
• Edge computing & resource allocation
• Deep Reinforcement Learning

• Email: pyeongjun.choi@dgist.ac.kr
• Office: No. 507B, Woojung Hall of Informatics, Korea University
• Google Scholar · ORCID · ICNL