The 17th Asian Conference on Machine Learning (ACML 2025) will take place between December 09-12, 2025, in Taipei, Taiwan. The conference aims to provide a leading international forum for researchers in machine learning and related fields to share their new ideas, progress and achievements.
Research papers from the Mathematical Optimization Laboratory (Professor IIDUKA Hideaki) at the Department of Computer Science, the School of Science and Technology has been accepted for presentation at ACML 2025.
The 17th Asian Conference on Machine Learning（ACML2025）
| Presenting author | IMAIZUMI Kento(2nd-year pre-doctoral student) |
|---|---|
| Title | Both Asymptotic and Non-Asymptotic Convergence of Quasi-Hyperbolic Momentum using Increasing Batch Size |
| Presenting author | OOWADA Kanata (2nd-year pre-doctoral student) |
|---|---|
| Title | Faster Convergence of Riemannian Stochastic Gradient Descent with Increasing Batch Size |
| Presenting author | KAMO Keisuke (2nd-year pre-doctoral student) |
|---|---|
| Title | Increasing Batch Size Improves Convergence of Stochastic Gradient Descent with Momentum |
≪Japanese version≫