Hello, Everyone!
I am a Ph.D. student in the Brain and Machine Intelligence Lab. (BMIL) at Sungkyunkwan University, Republic of Korea.
My research focuses on designing robust and scalable AI systems that can learn, adapt, and generalize under real-world constraints, including distributed environments (Federated learning, Split learning) and open-world scenarios (Post-training adaptation, Novel class discovery).
🔵 [Feb. 2026] My paper "Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift" has been accepted to the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)!
🔵 [Jan. 2026] My journal article "Evolving Intelligent Network Attack Classifier under Label Distribution Shift" has been accepted to the IEEE Transactions on Network Science and Engineering (IEEE TNSE) (Rating: Q1, Rank: 3/136-Top 1.8%, IF: 7.9)!
🟢 [Nov. 2025] I am honored to have been selected as a recipient of the AI SeoulTech Graduate Scholarship (KRW 10,000,000) from the Seoul Scholarship Foundation!
🔵 [Oct. 2025] My journal article "Contrastive Learning Based Network Attack Classifier for Imbalanced Data" has been accepted to the Journal of Communications and Networks (JCN)!
🔵 [Sep. 2025] My paper has been accepted at NeurIPS 2025 AI4NextG Workshop. See you in San Diego!
🔵 [Aug. 2025] My journal article "Personalized Split Federated Learning with Early-exit: Pre-training and Online Learning Against Label Shifts" has been accepted to the IEEE Internet of Things Journal (IEEE IoTJ) (Rating: Q1, Rank: 9/250-Top 3.4%, IF: 8.2)!
🔵 [Aug. 2025] My two papers have been accepted at ACM CIKM 2025. See you in Seoul, Korea!
🔵 [Apr. 2025] My paper has been accepted at ACM MobiSys 2025. See you in Anaheim!
🟢 [Nov. 2024] My work won the HFR paper competition (Awarded KRW 15,000,000)!
🟢 [Jun. 2024] I was awarded the Soongsil Alumni Association Scholarship (Awarded KRW 5,000,000) as one of only two selected Ph.D. students!
🔵 [Jun. 2023] My two papers have been accepted at ICML 2023 Workshops. See you in Hawaii!
🔵 [Aug. 2022] My journal article “Hierarchical Detection of Network Anomalies: A Self-supervised Learning Approach" has been accepted to IEEE Signal Processing Letters!
Sungkyunkwan University, Republic of Korea
[Ongoing] Ph.D. Student, Department of Electrical and Computer Engineering (Mar.2024~)
Advisor: Professor Minhae Kwon
Soongsil University, Republic of Korea
M.S., Department of Information and Telecommunication Engineering (Sep.2022-Feb.2024)
Thesis Title: Federated Clients and Cloud Collaboration System with Knowledge Augmentation
Advisor: Professor Minhae Kwon
B.S., School of Electronic Engineering (Mar.2019-Aug.2022) (Early Graduation)
Senior Project(Rank: 2/225)-Network Anomaly Detection System Using Hidden Layer Information of Autoencoder Based on RaPP
Language
Korean: Native
English: Fluent
Awards
1. Best Paper - Journal of Korean Institute of Communications and Information Sciences, Vol. 50, No. 12 (2025.12)
Title: Personalized Federated Post-training for Real-time Adaptation to Label Distribution Shifts
2. Best Paper Award - The 2nd HFR paper competition (2024.11) - Awarded KRW 15,000,000
Title: Online Label Shift Adaptation for Network Intrusion Detection Systems
3. Best Paper - Journal of Korean Institute of Communications and Information Sciences, Vol. 49, No. 7 (2024.07)
Title: Fine-tuning Anomaly Classifier for Unbalanced Network Data
4. Best Paper Award - 2023 Korean Institute of Communications and Information Sciences Summer Conference (2023.12.)
Title: Robust Partial Share Federated Learning Against Model Poisoning Attack
5. Best Paper Award -2022 Korean Institute of Communications and Information Sciences Winter Conference (2022.06.22)
Title: A Study on Detection Role of Hierarchical Stage in Autoencoder-Based Network Intrusion Detection Systems
6. Dean's Award (2nd Place) - 2021 Soongsil University School of Electronic Engineering Senior Project Competition (2021.11.05)
Title: Network Anomaly Detection System Using Hidden Layer Information of Autoencoder Based on RaPP
7. Participation Award - 2021 Korean Institute of Communications and Information Sciences ICT Idea Competition (2021.06.18)
Title: Regional Infection Risk Prediction System Using Population Flow and SIR Model Analysis by District
Scholarships and Travel Grants
1. AI SeoulTech Graduate Scholarship, Seoul Scholarship Foundation (2025.11.16) - Awarded KRW 10,000,000
2. Alumni Association Scholarship, Soongsil University (2024.06.18) - Awarded KRW 5,000,000
One of only two recipients selected among all Ph.D. students across departments for academic excellence and research potential
1. Brain and Machine Intelligence Lab. (BMIL)
Graduate Researcher (Aug. 2022 - Present, Advisor: Prof. Minhae Kwon)
Undergraduate Researcher (June. 2021 - Aug. 2022, Advisor: Prof. Minhae Kwon)
2. Smart City Building Idea Competition & Curriculum (KICS Winter Conference 2022, Pyeongchang, Republic of Korea)
3. AI Innovation Square (National IT Industry Promotion Agency (NIPA), Seoul, Republic of Korea)
Advanced Vision Course (2020.07.01 ~ 2020.08.31)
1. Teaching Assistant (AI for Multimedia Technologies), Soongsil University, School of Electronic Engineering (2025.03 ~ 2022.06)
2. Program Assistant (Global IT Training Program), Soongsil University, School of Electronic Engineering (2025.01)
Led undergraduate students on a global IT training program featuring CES 2025, Silicon Valley industry tours, and a Stanford University visit
3. Teaching Assistant (Discrete Mathematics), Soongsil University, School of Electronic Engineering (2022.03 ~ 2022.06)
4. Student Mentor, Soongsil University, School of Electronic Engineering(2022.03 ~ 2022.06)
Heewon Park (M.S. Student, BMIL, December 2022 ~ December 2025)
Mugon Joe (M.S. Student, BMIL, June 2023 ~ December 2025)
Kyungjin Im (M.S. Student, BMIL, December 2024 ~ )
International Journal Articles & Conference Papers
[12] Kyungjin Im, Chanin Eom, Miru Kim, Minhae Kwon, Anonymous submission – model merging, Top AI Conference, under review.
[11] Miru Kim, Heewon Park, Minhae Kwon, Anonymous submission – federated learning, Top AI Conference, under review.
[10] Heewon Park*, Mugon Joe*, Miru Kim*, Kyungjin Im, Minhae Kwon, "Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift," IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Accepted. (*equal contribution)
[9] Miru Kim*, Mugon Joe*, Minhae Kwon, "Evolving Intelligent Network Attack Classifier under Label Distribution Shift," IEEE Transactions on Network Science and Engineering (IEEE TNSE), Early access. (*equal contribution) [link] (Rating: Q1, Rank: 3/136-Top 1.8%, IF: 7.9)
[8] Mugon Joe*, Miru Kim*, Minhae Kwon, "Contrastive Learning Based Network Attack Classifier for Imbalanced Data," Journal of Communications and Networks (JCN), vol. 28, no. 1, Feb. 2026. (*equal contribution) [link]
[7] Miru Kim*, Heewon Park*, Minhae Kwon, “Personalized Split Federated Learning with Early-exit: Pre-training and Online Learning Against Label Shifts,” IEEE Internet of Things Journal (IEEE IoTJ), vol.12, no.22, Nov. 2025. (*equal contribution) [link] (Rating: Q1, Rank: 9/250-Top 3.4%, IF: 8.2)
[6] Miru Kim, Mugon Joe, Minhae Kwon, "OASIS: Open-world Adaptive Self-supervised and Imbalanced-aware System," ACM International Conference on Information and Knowledge Management (ACM CIKM), Nov. 2025. (BK21+ IF: 3, Acceptance rate: 26.86%) [link]
[5] Miru Kim, Mugon Joe, Minhae Kwon, "Improving Network Attack Classification on Imbalanced Real-world Intrusion Incident Datasets," ACM International Conference on Mobile Systems, Applications, and Services (ACM MobiSys), Jun. 2025. (BK21+ IF: 3) [link]
[4] Heewon Park, Mugon Joe, Miru Kim, Minhae Kwon, "ASAP: Unsupervised Post-training with Label Distribution Shift Adaptive Learning Rate," ACM International Conference on Information and Knowledge Management (ACM CIKM), Nov. 2025. (BK21+ IF: 3, Acceptance rate: 30.08%) [link]
[3] Heewon Park*, Miru Kim*, Minhae Kwon, "Personalized Federated Sensing for Heterogeneous Environment," IEEE Sensors Letters (IEEE SL), vol. 9, no. 4, 2024. (*equal contribution) [link]
[2] Hyoseon Kye, Miru Kim, Minhae Kwon, "Hierarchical Detection of Network Anomalies: A Self-supervised Learning Approach," IEEE Signal Processing Letters (IEEE SPL), vol. 29, pp. 1908-1912, 2022. [link][Newsis][AItimes]
[1] Hyoseon Kye, Miru Kim, Minhae Kwon, “Hierarchical Autoencoder for Network Intrusion Detection,” IEEE International Conference on Communications (IEEE ICC), May. 2022. [link][Media]
International Conference Workshop Papers
[3] Miru Kim, Minhae Kwon, "Network Traffic Foundation Model with Adaptation via In-Context Learning and Mixture-of-Experts," Conference on Neural Information Processing Systems (NeurIPS) AI4NextG Workshop, Dec 2025.
[2] Miru Kim, Minhae Kwon, “Augmenting the Knowledge to Large Model from Federated Small Models,” International Conference on Machine Learning (ICML) workshop, July 2023.
[1] Heewon Park, Miru Kim, Minhae Kwon, “Localizing Partial Model for Personalized Federated Learning,” International Conference on Machine Learning (ICML) workshop, July 2023.
Domestic Journal Articles
[4] Kyungjin Im, Heewon Park, Miru Kim, Mugon Joe, Minhae Kwon, “Personalized Federated Post-training for Real-time Adaptation to Label Distribution Shifts,” Journal of Korean Institute of Communications and Information Sciences (JKICS), vol.50, no.12, pp.1811-1821, Dec 2025. (Best Paper)
[3] Mugon Joe, Miru Kim, Minahe Kwon, "Fine-tuning Anomaly Classifier for Unbalanced Network Data," Journal of Korean Institute of Communications and Information Sciences (JKICS), vol. 49, no. 7, Jul. 2024. (Best Paper) [pdf]
[2] Heewon Park, Miru Kim, Minhae Kwon, "Robust Partial Share Federated Learning Algorithm against Model Poisoning Attack," Journal of Korean Institute of Communications and Information Sciences (JKICS), vol. 48, no. 11, Nov. 2023. [pdf]
[1] Miru Kim, Hyoseon Kye, Minhae Kwon, "Network Anomaly Detection System Using Hidden Layer Information of Autoencoder," Journal of Korean Institute of Communications and Information Sciences (JKICS), vol. 47, no. 9, Sep. 2022. [pdf]
Domestic Conference Papers
[17] Kyungjin Im, Heewon Park, Miru Kim, Mugon Joe, Minhae Kwon, “FIM-based Personalized Federated Post-training Solution Robust to Data Distribution Shift,” Korea Computer Congress, July 2025. (Best Paper Award) [Video]
[16] Kyungjin Im, Heewon Park, Miru Kim, Minhae Kwon, “Federated Post-training for Personalized Client Models,” the Joint Conference on Communications and Information, Jul. 2025. [Video]
[15] Miru Kim, Minhae Kwon, "xLSTM-based Autoencoder for Anomaly Detection of Time Series Network Traffic Data, " Korean Institute of Communications and Information Sciences Winter Conference, Feb. 2025. [pdf] [video]
[14] Miru Kim, Heewon Park, Minhae Kwon, "Federated Learning for Multiple SDVs and Cloud Collaborative Systems," Korean Society of Automotive Engineers Autumn Conference, Nov. 2024. [pdf] [video]
[13] Heewon Park, Miru Kim, Minhae Kwon, "Robust Federated Learning for SDV Against Adversarial Poisoning Attack," Korean Society of Automotive Engineers Autumn Conference, Nov. 2024. [pdf]
[12] Miru Kim, Heewon Park, Minhae Kwon "Online Learning for SDV and Cloud AI Models against Data Distribution Shifts," Korean Society of Automotive Engineers Autumn Conference, Nov. 2024. [pdf]
[11] Mugon Joe, Miru Kim, Minhae Kwon, "Contrastive Learning Based Adaptation Algorithms for CAN Intrusion Detection on Shifting Attack Types," Korean Society of Automotive Engineers Autumn Conference, Nov. 2024. [pdf]
[10] Mugon Joe, Miru Kim, Minhae Kwon, "Hierarchical Autoencoder Based Network Intrusion Detection Systems for CAN Communications," Korean Society of Automotive Engineers Autumn Conference, Nov. 2024. [pdf]
[9] Mugon Joe, Miru Kim, Minhae Kwon, "Fine-tuning Solution for Hierarchical Autoencoder Anomaly Classifiers," Korean Institute of Communications and Information Sciences Fall Conference, Nov. 2023. [pdf]
[8] Heewon Park, Miru Kim, Minhae Kwon, "Robust Partial Share Federated Learning Against Model Poisoning Attack," Korean Institute of Communications and Information Sciences Summer Conference, Jun. 2023. (Best Paper Award) [pdf]
[7] Heewon Park, Miru Kim, Minhae Kwon, "Performance Comparison of Partial Share Algorithms for Personalized Federated Learning," Joint Conference on Communications and Information, Apr. 2023. [pdf]
[6] Miru Kim, Minhae Kwon, "Performance Analysis of Partial-share Solution for Personalized Federated Learning," Korean Institute of Communications and Information Sciences Winter Conference, Feb. 2023. [pdf] [video]
[5] Miru Kim, Minhae Kwon, "Low-complex Anomaly Detection Method Using Important Features," Korea Artificial Intelligence Conference, Sep. 2022. [pdf] [video]
[4] Miru Kim, Hyoseon Kye, Sujin Ahn, Minhae Kwon, “Performance Comparison of Neural Network Models for Network Intrusion System,” Joint Conference on Communications and Information, Apr. 2022. [pdf] [video]
[3] Sujin Ahn, Hyoseon Kye, Miru Kim, Minhae Kwon, “Study on High-impact Features in Multiple Autoencoder Neural Network Models,” Joint Conference on Communications and Information, Apr. 2022. [pdf]
[2] Miru Kim, Hyoseon Kye, Minhae Kwon, “A Study on Detection Role of Hierarchical Stage in Autoencoder Based Network Intrusion Detection Systems,” Korean Institute of Communications and Information Sciences Winter Conference, Feb. 2022. (Best Paper Award) [pdf] [video]
[1] Miru Kim, Hyoseon Kye, Minhae Kwon, “RaPP-based Network Anomaly Detection Systems,” Korea Artificial Intelligence Conference, Sep. 2021. [pdf] [video]
US Patent Applications
1. Miru Kim, Minhae Kwon, “Method For Detecting Anomalies In A Vehicle Network, A Computing Device For Performing The Same, And A Vehicle Equipped With The Same,” US Patent, November 24, 2025. (Application no. US 19/398,210)
2. Miru Kim, Mugon Joe, Minhae Kwon, “Method For Classifying Anomaly Types In A Vehicle Network, Computing Device For Performing The Same, And Vehicle Equipped With The Same,” US Patent, November 24, 2025. (Application no. US 19/398,225)
3. Miru Kim, Heewon Park, Minhae Kwon, “Learning method and system based on federated learning and computing device for performing the same,” US Patent, July 14, 2025. (Application no. US 19/267,902)
4. Heewon Park, Miru Kim, Minhae Kwon, “Split federated learning system and method optimized for various hardware environments and local client for performing the same,” US Patent, July 14, 2025. (Application no. US 19/267,897)
5. Mugon Joe, Miru Kim, Minhae Kwon, “Method for Training Classification Model and Computing Device for Performing Same,” US Patent, February 7, 2025. (Application no. US 19/048,192)
6. Mugon Joe, Miru Kim, Minhae Kwon, “Fine-tuning Method and System for Anomaly Type Classification,” US Patent, July 23, 2024. (Application no. US 18/781,284)
7. Miru Kim, Heewon Park, Minhae Kwon, “Method and Apparatus for Augmenting Knowledge Using Information of Federated Learning,” US Patent, Jan 4, 2024. (Application no. US 18/404143)
8. Miru Kim, Minhae Kwon, "Apparatus And Method Of Data Anomaly Detection Based On Important Feature Value And Low Complexity Model," US Patent, July 7, 2023 (Application no. US 18/219642)
9. Miru Kim, Hyoseon Kye, Minhae Kwon, "Apparatus And Method Of Personalized Federated Learning Based On Partial Parameters Sharing," US Patent, July 7, 2023 (Application no. US 18/219691)
10. Hyoseon Kye, Miru Kim, Minhae Kwon, "System for Detecting Hierarchical Network Intrusion Using Hidden Layer Information of Autoencoder and Method Thereof," US Patent, November 2, 2022 (Application no. US 17/979728)
1. 김미르, 계효선, 권민혜, "오토인코더의 은닉층 정보를 이용한 계층적 네트워크 침입탐지 시스템 및 그 방법," Korean Patent, April 28, 2021 (Patent no. 10-2642060 )
1. 김미르, 조무곤, 권민혜, “데이터 분포 변화 및 새로운 클래스에 대해 적응적으로 분류할 수 있는 분류 모델의 학습 방법 및 이를 수행하기 위한 컴퓨팅 장치,” Korean Patent, filed on September 18, 2025 (Application no. 10-2025-0134733)
2. 박희원, 조무곤, 김미르, 권민혜, “데이터 분포 변화 추정 기반의 연합 학습 방법 및 시스템과 이를 수행하기 위한 컴퓨팅 장치,” Korean Patent, filed on September 16, 2025 (Application no. 10-2025-0132770)
3. 박희원, 조무곤, 김미르, 권민혜, “실시간 환경에서 데이터 분포 변화를 이용한 인공지능 모델의 사후 학습 방법 및 이를 수행하기 위한 컴퓨팅 장치,” Korean Patent, filed on September 16, 2025 (Application no. 10-2025-0132771)
4. 김미르, 권민혜, “차량 네트워크의 이상 탐지 방법 및 이를 수행하기 위한 컴퓨팅 장치와 이를 구비하는 차량,” Korean Patent, filed on September 11, 2025 (Application no. 10-2025-0129711)
5. 김미르, 조무곤, 권민혜, “차량 네트워크의 이상 유형 분류 방법 및 이를 수행하기 위한 컴퓨팅 장치와 이를 구비하는 차량,” Korean Patent, filed on September 10, 2025 (Application no. 10-2025-0129416)
6. 임경진, 박희원, 김미르, 권민혜, “연합 학습 기반의 사후 학습을 위한 학습 방법 및 시스템과 이를 수행하기 위한 컴퓨팅 장치,” Korean Patent, filed on September 23, 2025 (Application no. 10-2025-0137594)
7. 김미르, 박희원, 권민혜, " 연합 학습 기반의 학습 방법 및 시스템과 이를 수행하기 위한 컴퓨팅 장치," Korean Patent, filed on April 21, 2025 (Application no. 10-2025-0051898)
8. 김미르, 박희원, 권민혜, "다양한 하드웨어 환경에 최적화된 분할 연합 학습 시스템 및 방법과 이를 수행하기 위한 로컬 클라이언트," Korean Patent, filed on April 08, 2025 (Application no. 10-2025-0045651)
9. 조무곤, 김미르, 권민혜, “분류 모델의 학습 방법 및 이를 수행하기 위한 컴퓨팅 장치,” Korean Patent, filed on November 15, 2024 (Application no. 10-2024-0162776)
10. 조무곤, 김미르, 권민혜, “분류 모델의 적응 학습 방법 및 장치,” Korean Patent, filed on November 15, 2024 (Application no. 10-2024-0162777)
11. 김미르, 조무곤, 권민혜, “이상유형 분류를 위한 미세조정 방법 및 시스템,” Korean Patent, filed on April 11, 2024 (Application no. 10-2024-0048667)
12. 김미르, 박희원, 권민혜, “연합학습 정보를 이용한 지식 증강 방법 및 장치,” Korean Patent, filed on September 7, 2023 (Application no. 10-2023-0119121)
13. 김미르, 계효선, 권민혜, "파라미터 부분 공유에 기초한 개인화 연합 학습 방법 및 장치," Korean Patent, January 17, 2023 (Application no. 10-2023-0007036)
14. 김미르, 권민혜, "중요 특성 값 및 저복잡 모델에 기초한 데이터 이상 탐지 방법 및 장치," Korean patent, January 17, 2023 (Application no. 10-2023-0007035)