Embedded Files
I am Hyunju Kim, a Postdoctoral Researcher at CDSN LAB@KAIST.
I am Hyunju Kim, a Postdoctoral Researcher at CDSN LAB@KAIST.
My research lies at the intersection of Human Activity Recognition (HAR), Machine Learning, and Ubiquitous Computing, with a particular focus on developing cutting-edge methods for Self-Learning Activity Recognition and Personalization Systems in the open world.
I am deeply fascinated by the challeng of time series analysis, sensor data fusion, temporal relation modeling, and causality-aware pattern mining, which are critical for advancing the effectiveness of activity recognition systems. Furthermore, I am keenly interested in the integration of smart devices, ambient sensors, wearable sensors, and robotics to develop personalized and adaptive service solutions, utilizing reinforcement learning to drive context-aware and user-specific automation systems.
Currently, my work explores the application of deep learning techniques such as Transformers and contrastive learning for generalizable novelty detection and early recognition of human activities, aiming to enhance user experience and service optimization in intelligent environments.
Through these research endeavors, I aim to contribute to the development of more intuitive, responsive, and human-centric smart environments that seamlessly integrate into everyday life.
News
News
2025.08: Paper accepted to ACM CIKM 2025.
2025.06: Selected as an Outstanding Reviewer for KDD 2025 (Feb Cycle).
2024.12: Successfully defended Ph.D. Thesis (A context-aware human activity recognition architecture for ambient sensor-based smart environments).
2024.12: Selected as an Excellent Reviewer for KDD 2025 (Aug Cycle).
2024.11: Awarded the best paper award from KAIST.
2024.07: Paper published in IEEE COMPSAC 2024.
2024.05: Paper published in Nature Scientific Data 2024.
Publications
Publications
Hyunju Kim and Dongman Lee, "Self-supervised Dual-view Framework with Tailored Negative Sampling for New Activity Detection", CIKM '25 (to appear)
Hyunju Kim, Heesuk Son, and Dongman Lee, "Causality-aware Pattern Mining Scheme for Group Activity Recognition", COMPSAC '24
Hyunju Kim*, Geon Kim*, Akanksha Jain, and Dongman Lee, "Mobile Robot based Personalized Thermal Comfort Scheme", UR '24
Hyunju Kim, Geon Kim, Taehoon Lee, Kisoo Kim, and Dongman Lee, "A Dataset of Ambient Sensors in a Meeting Room for Activity Recognition", Scientific Data 11, no. 1 (May 2024).
Geon Kim, Hyunju Kim, Youngjae Kim, and Dongman Lee, "Exploring Technical Challenges for Thermal Sensation Estimation in Mobile Robot Driven Personalized Thermal Control", UR '23
Geon Kim, Hyunju Kim, and Dongman Lee, "Towards Deployment of Mobile Robot driven Preference Learning for User-State-Specific Thermal Control in a Real-World Smart Space", SAC '23
Tae-Hoon Lee, Hyunju Kim, and Dongman Lee, "Transformer Based Early Classification for Real-Time Human Activity Recognition in Smart Homes", SAC '23
Kisoo Kim, Hyunju Kim, and Dongman Lee, "A Correlation-based Real-time Segmentation Scheme for Multi-user Collaborative Activities", COMPSAC '22
Hyunju Kim and Dongman Lee, "AR-T: Temporal Relation embedded Transformer for the Real World Activity Recognition", MobiQuitous '21
Hyunju Kim and Dongman Lee, "TAP: A Transformer based Activity Prediction Exploiting Temporal Relations in Collaborative Tasks", PerCom '21 Workshops
Hyunju Kim and Dongman Lee, "ADELA: Attention based Deep Ensemble Learning for Activity Recognition in Smart Collaborative Environments", SAC '21
Heesuk Son, Jeongwook Park, Hyunju Kim, and Dongman Lee, "Distributed Multi-agent Preference Learning for an IoT-enriched Smart Space", ICDCS '19
Education
Education
2018 – 2024: PhD Degree in School of Computing, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea
2015 – 2017: Master Degree in School of Computing, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea
2012 – 2015: Bachelor Degree in Computer Engineering, Ajou University, Suwon, South Korea, GPA : 4.31/4.5 (graduated in the top 5%), Early Graduation
Academic Services
Academic Services
Outstanding Reviewer, KDD 2025 Feb Cycle
Excellent Reviewer, KDD 2025 Aug Cycle
Reviewer, Knowledge-Based Systems 2024
Page updated
Google Sites
Report abuse