Kwang-Sung Jun

Interests

Teaching

Machine learning, theory of machine learning, online learning (optimization), multi-armed bandits.

Research

interactive machine learning, multi-armed bandits, online learning, active learning

Courses

Scholarly Contributions

Journals/Publications

• Orabona, F., & Jun, K. (2021). Tight Concentrations and Confidence Sequences from the Regret of Universal Portfolio. arXiv preprint arXiv:2110.14099.

Proceedings Publications

• Bian, J., & Jun, K. (2022). Maillard Sampling: Boltzmann Exploration Done Optimally. In Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS).
  More info

  CORE A
• Faury, L., Abeille, M., Jun, K., & Calauz`{e}nes, C. (2022). Jointly Efficient and Optimal Algorithms for Logistic Bandits. In Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS).
  More info

  core A
• Gales, S. B., Sethuraman, S., & Jun, K. (2022). Norm-Agnostic Linear Bandits. In Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS).
  More info

  core A
• Kim, Y., Yang, I., & Jun, K. (2022). Improved regret analysis for variance-adaptive linear bandits and horizon-free linear mixture mdps. In submitted to Proceedings of the International Conference on Machine Learning (ICML).
• Mason, B., Jun, K., & Jain, L. (2022). An Experimental Design Approach for Regret Minimization in Logistic Bandits. In AAAI Conference on Artificial Intelligence (AAAI).
  More info

  core A*
• Jang, K., Jun, K., Yun, S. Y., & Kang, W. (2021). Improved Regret Bounds of Bilinear Bandits using Action Space Dimension Analysis. In Proceedings of the International Conference on Machine Learning (ICML).
  More info

  core A*
• Jun, K., Jain, L., Mason, B., & Nassif, H. (2021). Improved Confidence Bounds for the Linear Logistic Model and Applications to Linear Bandits. In Proceedings of the International Conference on Machine Learning (ICML).
  More info

  core A*
• Park, H., Shin, S., Jun, K., & Ok, J. (2021). Transfer Learning in Bandits with Latent Continuity. In IEEE International Symposium on Information Theory (ISIT).
  More info

  CORE B
• Zhang, C., Jun, K., Jun, K., & Zhang, C. (2020, December). Crush Optimism with Pessimism: Structured Bandits Beyond Asymptotic Optimality.. In Neural Information Processing Systems.
  More info

  NeurIPS is a top-tier conference (A* in CORE).
• Jun, K., & Orabona, F. (2019, Dec). Parameter-Free Locally Differentially Private Stochastic Subgradient Descent. In NeurIPS Workshop on Privacy in Machine Learning (PriML).
• Jun, K., & Orabona, F. (2019, Jun). Parameter-Free Online Convex Optimization with Sub-Exponential Noise. In Proceedings of the Conference on Learning Theory, 99.
  More info

  COLT is a top-tier conference (A* in CORE).
• Jun, K., Cutkosky, A., & Orabona, F. (2019, Dec). Kernel Truncated Randomized Ridge Regression: Optimal Rates and Low Noise Acceleration. In Advances in Neural Information Processing Systems.
  More info

  NeurIPS is a top-tier conference (A* in CORE).
• Jun, K., Willett, R., Wright, S., & Nowak, R. (2019, Jun). Bilinear Bandits with Low-rank Structure. In Proceedings of the 36th International Conference on Machine Learning, 97.
  More info

  ICML is a top-tier conference (A* in CORE).

Others

• Jun, K., & Zhang, C. (2020, July). Crush Optimism with Pessimism: Structured Bandits Beyond Asymptotic Optimality.. ICML Workshop on Theoretical Foundations of Reinforcement Learning.
  More info

  This was a preliminary work that was presented and published in ICML Workshop on Theoretical Foundations of Reinforcement Learning.