Showing stat.MLShow all
2 papers · 1 filter
stat.ML2026
Momentum in large-batch training: Polyak enlarges the critical batch size, Nesterov improves data efficiency
Jia-Nan Wang, Zixun Huang, Kairui Li +2
We study when and how momentum improves large-batch training in the one-pass regime, using power-law kernel regression as a tractable setting. We first characterize stochastic stab…
stat.ML2025
Variance-Aware Baselines and Adaptive Learning Rates for Reinforcement Learning with Verifiable Rewards
Zixun Huang, Jiayi Sheng, Zeyu Zheng
Reinforcement learning with verifiable rewards (RLVR) has emerged as an effective paradigm for post-training large language models, yet the design of its baselines and learning-rat…