Pith. sign in

REVIEW 1 cited by

On the Parameterization of Second-Order Optimization Effective Towards the Infinite Width

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2312.12226 v2 pith:XIFM6TTL submitted 2023-12-19 cs.LG

classification cs.LG
keywords optimizationparameterizationsecond-orderlearningfeaturehyperparametersmodelsscales
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Second-order optimization has been developed to accelerate the training of deep neural networks and it is being applied to increasingly larger-scale models. In this study, towards training on further larger scales, we identify a specific parameterization for second-order optimization that promotes feature learning in a stable manner even if the network width increases significantly. Inspired by a maximal update parameterization, we consider a one-step update of the gradient and reveal the appropriate scales of hyperparameters including random initialization, learning rates, and damping terms. Our approach covers two major second-order optimization algorithms, K-FAC and Shampoo, and we demonstrate that our parameterization achieves higher generalization performance in feature learning. In particular, it enables us to transfer the hyperparameters across models with different widths.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Pre-Training LLMs on a budget: A comparison of three optimizers

    cs.LG 2025-07 conditional novelty 6.0 of 10

    In budget-constrained 2.7B-parameter LLM pre-training, Lion is fastest, Sophia reaches the lowest loss, but AdamW wins on downstream benchmarks.

Pith tools