Won Joon Yun
ywjoon95@korea.ac.kr
CV / Github / Scholar
About
I am Won Joon Yun, an alumni of the School of Electrical Engineering at Korea University.
Research Interest
My research interest is driven by the quest to optimize applications — reducing delay, minimizing both computing and communication resource consumption, and enhancing efficiency. This pursuit centers on computational and communication network challenges and extends into the realm of quantum computing, which stands at the forefront of next-generation computational and communicative resources. Among the numerous facets of quantum computing, I am especially intrigued by the process of problem definition and resolution, even in the face of current challenges in applying these technologies to real-world contexts. As both a researcher and an engineer, my goal is to bridge the divide between the idealized Hilbert space and the constraints of modern quantum computers. I aim to develop scalable quantum machine learning algorithms that consistently showcase quantum advantage and to be at the forefront of translating Quantum AI into tangible real-world applications.
Publications
Google Scholar provides a full list under chronological/citations order.
Selected publications are labeled by 🎯.
» Quantum
Pure QML
🎯Quantum Federated Learning with Entanglement Controlled Circuits and Superposition Coding
W. J. Yun, H. Baek, J. P. Kim, S. Jung, J. Park, M. Bennis and J. Kim
arXiv:2212.01732. [Link]
[TL; DR] Depth-controllable U3 circuit in federated learning framework can achieve better performance with fidelity regularizer.🎯Quantum Multi-Agent Reinforcement Learning via Variational Quantum Circuit Design
W. J. Yun, Y. Kwak, J.P. Kim, H. Lee, S. Jung, J. Park and J. Kim,
Proceedings of IEEE International Conference on Distributed Computing Systems, July 2022. [Link][Slide][Code]
[TL;DR] We introduce the first quantum reinforcement learning framework in multi-agent settings.🎯Quantum Multi-Agent Meta Reinforcement Learning
W. J. Yun, J. Park, and J. Kim
Proceedings of AAAI Conference on Artificial Intelligence (Main Track), February 2023. [Link][Slide][Youtube]
[TL; DR] The trainable measurement in multi-agent system can achieve Pareto-optimal from Nash equilibrium by only changing the measurement origin.🎯Projection Valued Measure-based Quantum Machine Learning for Multi-Class Classification
W. J. Yun, H. Baek, and J. Kim
arXiv:2210.16731. [Link]
[TL; DR] Measuring quantum state to all possible events, QNN can achieve great performances in MNIST/FashionMNIST/CIFAR10/EMNIST.Slimmable Quantum Federated Learning
W. J. Yun, J. P. Kim, S. Jung, J. Park, M. Bennis, and J. Kim
ICML Workshop on Dynamic Neural Networks, July 2022. (Spotlight Oral Presentation) [Link][Slide]
[TL; DR] Trainable measurement with superposition coding and successive decoding improves communication opportunity in federated learning settings.Stereoscopic Scalable Quantum Convolutional Neural Networks
H. Baek, W. J. Yun, S. Park, and J. Kim
Neural Networks (Elsevier), in Press. [Link]
[TL; DR] We firstly introduce 3D quantum convolutional neural network and regularizer technique, which can process voxel (Extended version).3D Scalable Quantum Convolutional Neural Networks for Point Cloud Data Processing in Classification Applications
H. Baek, W. J. Yun, J. Kim
AAAI Conference on Artificial Intelligence Workshops on AI2ASE, February 2023. [Link]
[TL; DR] We firstly introduce 3D quantum convolutional neural network and regularizer technique, which can process voxel.Scalable Quantum Convolutional Neural Networks
H. Baek, W. J. Yun, J. Kim
arXiv:2209.12372. [Link]
[TL; DR] We introduce multiple ansatz-based quantum-classical hybrid convolutional neural network and regularizer for feature diversity.FV-Train: Quantum Convolutional Neural Network Training with a Finite Number of Qubits by Extracting Diverse Features (Student Abstract)
H. Baek, W. J. Yun, J. Kim
Proceedings of AAAI Conference on Artificial Intelligence, February 2023. (Top 25 Finalist) [Link]
[TL; DR] We introduce inverse-fidelity regularizer to guarantee feature diversity made by quantum neural networks.Quantum Split Neural Network Learning using Cross-Channel Pooling
W. J. Yun, H. Baek, and J. Kim
arXiv:2211.06524. [Link]
[TL; DR] Quantum-version of split learning with cross-channel pooling can improve reducing the communication cost and accuracy.
Tutorial/Survey
Quantum Distributed Deep Learning Architectures: Models, Discussions, and Applications
Y. Kwak, W. J. Yun (co-first author), J. P. Kim, H. Cho, J. Park, M. Choi, S. Jung, and J. Kim
ICT Express (Elsevier), 9(5):486-491, October 2023. [Link]Introduction to Quantum Reinforcement Learning: Theory and PennyLane-based Implementation
Y. Kwak, W. J. Yun, S. Jung, J. -K. Kim, and J. Kim
Proceedings of IEEE International Conference on ICT Convergence, October 2021. [Link]Quantum Neural Networks: Concepts, Applications, and Challenges
Y. Kwak, W. J. Yun, S. Jung, and J. Kim,
Proceedings of IEEE International Conference on Ubiquitous and Future Networks, August 2021.[Link]
Applications
Quantum Multi-Agent Actor-Critic Networks for Cooperative Mobile Access in Multi-UAV Systems
C. Park, W. J. Yun, J. P. Kim, T. K. Rodrigues, S. Park, S. Jung and J. Kim
IEEE Internet of Things Journal. [Link]
[TL; DR] We implemented quantum multi-agent reinforcement learning in multi-UAV system, which shown robustness compared to classical ML.Quantum Multi-Agent Actor-Critic Neural Networks for Internet-Connected Multi-Robot Coordination in Smart Factory Management
W. J. Yun, J.P. Kim, S. Jung, J.-H. Kim, and J. Kim
IEEE Internet of Things Journal. [Link]
[TL; DR] We implemented quantum multi-agent reinforcement learning in multi-robot system, which shown robustness compared to classical ML.Visual Simulation Software Demonstration for Quantum Multi-Drone Reinforcement Learning
C. Park, J. P. Kim,W. J. Yun, S. Jung, J. Kim
arXiv:2211.15375. [Link]
[TL; DR] We introduce visual simulation of quantum multi-agent reinforcement learning in multi-UAV system.
» Classical AI
Network/Communications
🎯SlimFL: Federated Learning with Superposition Coding over Slimmable Neural Networks
W. J. Yun, Y. Kwak, H. Baek, S. Jung, M. Ji, M. Bennis, and J. Park, J. Kim
IEEE/ACM Transactions on Networking. [Link]
[TL; DR] With leveraging (1) slimmable neural networks, (2) superposition coding & successive decoding, we achieve computing/communication efficiency.
(Extended version of INFOCOM 2022 and ICML-FL'21)🎯 Joint Superposition Coding and Training for Federated Learning over Multi-Width Neural Networks
H. Baek, W. J. Yun (co-first author), Y. Kwak, S. Jung, M. Ji, M. Bennis, J. Park, and J. Kim
Proceedings of IEEE International Conference on Computer Communications (INFOCOM), May 2022. [Link]
[TL; DR] With leveraging (1) slimmable neural networks, (2) superposition coding & successive decoding, we achieve computing/communication efficiency.🎯 Attention-based Reinforcement Learning for Real-Time UAV Semantic Communication
W. J. Yun, B. Lim, S. Jung, Y. -C Ko, J. Park, J. Kim and M. Bennis
Proceedings of IEEE International Symposium on Wireless Communication Systems (ISWCS), September 2021. [Link]
[TL; DR] The emergent of machine language for coordination can be generated in multi-agent system.Joint Pilot Design and Channel Estimation using Deep Residual Learning for Multi-Cell Massive MIMO under Hardware Impairments
B. Lim, W. J. Yun, J. Kim, and Y. -C. Ko
IEEE Transactions on Vehicular Technology, 71(7):7599–7612, July 2022. [Link]
[TL; DR] Designing neural network to generate pilot and denoise hardware impairment, we improve SNR, and rate.Joint User Clustering, Beamforming, and Power Allocation for mmWave-NOMA with Imperfect SIC
B. Lim, W. J. Yun, J. Kim, and Y. Ko
IEEE Transactions on Wireless Communications, in Press.
[TL; DR] We leverage cross-entropy optimization for better suboptimal allocation in mmWave-NOMA with imperfect SIC.Communication and Energy Efficient Slimmable Federated Learning via Superposition Coding and Successive Decoding
H. Baek, W. J. Yun (co-first author), S. Jung, M. Ji, J. Park, J. Kim, and M. Bennis
ICML Workshop on Federated Learning for User Privacy and Data Confidentiality, July 2021.[Link]Quality-Aware Deep Reinforcement Learning for Streaming in Infrastructure-Assisted Connected Vehicles
W. J. Yun, D. Kwon, M. Choi, J. Kim, G. Caire, and A. F. Molisch
IEEE Transactions on Vehicular Technology, 71(2):2002–2017, February 2022.[Link]Delay-Sensitive and Power-Efficient Quality Control of Dynamic Video Streaming using Adaptive Super-Resolution
M. Choi, W. J. Yun, and J. Kim
arXiv:2110.05783. [Link]Cooperative Video Quality Adaptation for Delay-Sensitive Dynamic Streaming using Adaptive Super-Resolution
M. Choi, W. J. Yun, and J. Kim
Proceedings of IEEE WiOpt Workshop on Caching, Computing and Delivery in Wireless Networks, September 2022.[Link]
Automation/Control
Self-Configurable Stabilized Real-Time Detection Learning for Autonomous Driving Applications
W. J. Yun, S. Park, D. Mohaisen and J. Kim
IEEE Transactions on Intelligent Transportation Systems, Early Access.[Link]Parallelized and Randomized Adversarial Imitation Learning for Safety-Critical Self-Driving Vehicles
W. J. Yun, M. Shin, S. Jung, and J. Kim
IEEE/KICS Journal Communications Networks, 24(6):710–721, December 2022. [Link]Hierarchical Reinforcement Learning using Gaussian Random Trajectory Generation in Autonomous Furniture Assembly
W. J. Yun, D. Mohaisen, S. Jung, J. -K. Kim, and J. Kim
Proceedings of ACM International Conference on Information and Knowledge Management (CIKM), October 2022. [Link]Multi-Agent Deep Reinforcement Learning using Attentive Graph Neural Architectures for Real-Time Strategy Games
W. J. Yun, S. Yi and J. Kim
Proceedings of IEEE International Conference on Systems, Man, and Cybernetics (SMC), October 2021. [Link]
Biomedicine
Hierarchical Deep Reinforcement Learning-based Propofol Infusion Assistant Framework in Anesthesia
W. J. Yun, M. Shin, D. Mohaisen, K. Lee, and J. Kim
IEEE Transactions on Neural Networks and Learning Systems, Early Access.[Link]Deep Reinforcement Learning-based Propofol Infusion with a 3,000-subject Dataset in Anesthesia
W. J. Yun, M. Shin, S. Jung, J. Ko, H. -C. Lee, and J. Kim
Computers in Biology and Medicine (Elsevier), 156:106739, April 2023. [Link]Demo: Multi-Site Clinical Federated Learning using NLP Models and NVFlare
W. J. Yun, S. Kim, J. Kim
Proceedings of IEEE International Conference on Distributed Computing Systems, July 2023.
ICT Applications
🎯 Cooperative Multi-Agent Deep Reinforcement Learning for Reliable Surveillance via Autonomous Multi-UAV Control
W. J. Yun, S. Park, J. Kim, M. Shin, S. Jung, D. Mohaisen, and J.-H. Kim
IEEE Transactions on Industrial Informatics, 18(10):7086–7096, October 2022. [Link]Cooperative Multi-Agent Deep Reinforcement Learning for Reliable and Energy-Efficient Mobile Access via Multi-UAV Control
C. Park, H. Lee, W. J. Yun, S. Jung, C. Cordeiro, and J. Kim
arXiv:2210.00945. [Link]DDPG-based Deep Reinforcement Learning for Loitering Munition Mobility Control: Algorithm Design and Visualization
H. Lee, W. J. Yun, S. Jung, J.-H. Kim, and J. Kim
Proceedings of IEEE VTS APWCS, August 2022.[Link]Autonomous Aerial Mobility Learning for Drone-Taxi Flight Control
W. J. Yun, Y. J. Ha, S. Jung, and J. Kim
Proceedings of IEEE International Conference on ICT Convergence, October 2021. [Link]Orchestrated Scheduling and Multi-Agent Deep Reinforcement Learning for Cloud-Assisted Multi-UAV Charging Systems
S. Jung, W. J. Yun, M. Shin, J. Kim, and J. -H. Kim,
IEEE Transactions on Vehicular Technology, 70(6):5362–5377, June 2021. [Link]Visualization of Deep Reinforcement Autonomous Aerial Mobility Learning Simulations
G. Lee,W. J. Yun, S. Jung, J. Kim and J. -H. Kim
Proceedings of IEEE International Conference on Computer Communications (INFOCOM Demo), May 2021. [Link]Distributed deep reinforcement learning for autonomous aerial eVTOL mobility in drone taxi applications
W. J. Yun, S. Jung, J. -H. Kim, J. Kim
ICT Express (Elsevier), 7(1):1–4, March 2021. [Link]Infrastructure-Assisted Cooperative Multi-UAV Deep Reinforcement Energy Trading Learning for Big-Data Processing
S. Jung, W. J. Yun, J. Kim, and J. -H. Kim
IEEE International Conference on Information Networking, January 2021. (Best Paper Award) [Link]