Pith. sign in

Surge Phenomenon in Optimal Learning Rate and Batch Size Scaling

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

In current deep learning tasks, Adam style optimizers such as Adam, Adagrad, RMSProp, Adafactor, and Lion have been widely used as alternatives to SGD style optimizers. These optimizers typically update model parameters using the sign of gradients, resulting in more stable convergence curves. The learning rate and the batch size are the most critical hyperparameters for optimizers, which require careful tuning to enable effective convergence. Previous research has shown that the optimal learning rate increases linearly or follows similar rules with batch size for SGD style optimizers. However, this conclusion is not applicable to Adam style optimizers. In this paper, we elucidate the connection between optimal learning rates and batch sizes for Adam style optimizers through both theoretical analysis and extensive experiments. First, we raise the scaling law between batch sizes and optimal learning rates in the sign of gradient case, in which we prove that the optimal learning rate first rises and then falls as the batch size increases. Moreover, the peak value of the surge will gradually move toward the larger batch size as training progresses. Second, we conducted experiments on various CV and NLP tasks and verified the correctness of the scaling law.

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

Decomposing Prediction Mechanisms for In-Context Recall

cs.LG · 2025-07-02 · conditional · novelty 7.0

In a toy in-context recall task, label-based task initiation and observation-based continuation are distinct mechanisms with separate emergence times, and the same first-token versus second-token gap appears in an OLMo translation task.

citing papers explorer

Showing 1 of 1 citing paper.

  • Decomposing Prediction Mechanisms for In-Context Recall cs.LG · 2025-07-02 · conditional · none · ref 31 · internal anchor

    In a toy in-context recall task, label-based task initiation and observation-based continuation are distinct mechanisms with separate emergence times, and the same first-token versus second-token gap appears in an OLMo translation task.