paper-with-me

홈 › Papers

Meta-Principled Family of Hyperparameter Scaling Strategies

2022-10-10 · Sho Yaida

In this note, we first derive a one-parameter family of hyperparameter scaling strategies that interpolates between the neural-tangent scaling and mean-field/maximal-update scaling. We then calculate the scalings of dynamical observables -- network outputs, neural tangent kernels, and differentials of neural tangent kernels -- for wide and deep neural networks. These calculations in turn reveal a proper way to scale depth with width such that resultant large-scale models maintain their representation-learning ability. Finally, we observe that various infinite-width limits examined in the literature correspond to the distinct corners of the interconnected web spanned by effective theories for finite-width neural networks, with their training dynamics ranging from being weakly-coupled to being strongly-coupled.

📄 PDF Abstract BibTeX arXiv:2210.04909

Code (0)

등록된 구현이 없습니다.

Tasks

Representation Learning

Similar Papers 제목 키워드 기반

Deriving Hyperparameter Scaling Laws via Modern Optimization Theory

2026-03-16 · Egor Shulgin, Dimitri von Rütte, Tianyue H. Zhang, Niccolò Ajroldi 외 arxiv

Hyperparameter transfer has become an important component of modern large-scale training recipes. Existing methods, such as muP, primarily focus on transfer between model sizes, with transfer across batch sizes and train…

$μ$pscaling small models: Principled warm starts and hyperparameter transfer

2026-02-11 · Yuxin Ma, Nan Chen, Mateo Díaz, Soufiane Hayou 외 arxiv

Modern large-scale neural networks are often trained and released in multiple sizes to accommodate diverse inference budgets. To improve efficiency, recent work has explored model upscaling: initializing larger models fr…

Natural Evolution Strategies

2011-06-22 · Daan Wierstra, Tom Schaul, Tobias Glasmachers, Yi Sun 외

This paper presents Natural Evolution Strategies (NES), a recent family of algorithms that constitute a more principled approach to black-box optimization than established evolutionary algorithms. NES maintains a paramet…

Evolutionary Algorithmsglobal-optimization

EvoGrad: Efficient Gradient-Based Meta-Learning and Hyperparameter Optimization

2021-06-19 · NeurIPS 2021 12 · Ondrej Bohdal, Yongxin Yang, Timothy Hospedales

Gradient-based meta-learning and hyperparameter optimization have seen significant progress recently, enabling practical end-to-end training of neural networks together with many hyperparameters. Nevertheless, existing a…

Cross-Domain Few-Shotcross-domain few-shot learningFew-Shot LearningHyperparameter Optimization+1

Predictable Scaling Laws of Optimal Hyperparameters for LLM Continued Pre-training

2026-06-04 · Yongwei Zhou, Juncheng Diao, Junlin Shang, Peiguang Li 외 arxiv

The efficacy of continued pre-training for Large Language Models (LLMs) hinges upon hyperparameter configurations, such as learning rate and batch size. However, current practices often rely on heuristics or grid searche…