About Me

I am a Ph.D. student in the integrated M.S./Ph.D. program at KAIST AI, working in the Optimization & Machine Learning Laboratory (OptiML) under the supervision of Prof. Chulhee Yun. My research focuses on theoretical machine learning and optimization, particularly on understanding learning and optimization algorithms. I received my B.S. in Mathematical Sciences from KAIST.

News

Sep. 2026

Our paper, Beyond Rate Optimality: How Polyak’s Momentum Shapes Non-Convex Optimization, was accepted to the NeurIPS 2026 OPT Workshop.

May 2026

Our paper, Understanding Polyak’s Momentum in Deep Learning May Require Rethinking Non-Convex Optimization, was accepted to the ICML 2026 High-dimensional Learning Dynamics (HiLD) Workshop.

Mar. 2026

Transitioned to the integrated M.S./Ph.D. program in AI at KAIST.

May 2025

Our paper, Provable Benefit of Random Permutations over Uniform Sampling in Stochastic Coordinate Descent, was accepted to ICML 2025.

Feb. 2025

Started the M.S. program in AI at KAIST.

Publications

  • Beyond Rate Optimality: How Polyak's Momentum Shapes Non-Convex Optimization OpenReview (HiLD)
  • Donghwa Kim, Chulhee Yun

    NeurIPS 2026 Workshop on Optimization for Machine Learning (OPT)

    ICML 2026 Workshop on High-dimensional Learning Dynamics (HiLD)

  • Provable Benefit of Random Permutations over Uniform Sampling in Stochastic Coordinate Descent arXiv OpenReview
  • Donghwa Kim, Jaewook Lee, Chulhee Yun

    ICML 2025

  • Deep Symbolic Learning for Histogram-Valued Regression Data DOI
  • Ilsuk Kang, Donghwa Kim, Hosik Choi, Young Joo Yoon, Cheolwoo Park

    Statistical Analysis and Data Mining, 2025

    Academic Services

    Conference Reviewer

    • International Conference on Learning Representations (ICLR), 2027
    • Neural Information Processing Systems (NeurIPS), 2026