脑袋尖尖

PODCAST · technology

脑袋尖尖

脑袋尖尖,我们是伦敦的理工学生,想要做一个分享科技新闻,探索前沿科技和学术论文的播客。

  1. 1

    EP01 论文速读 | 简单反思《语言模型的 Scaling Law》,现在还适用吗?

    概述本期主题:深入解析 Scaling Law及其在模型性能优化中的应用。核心问题:Scaling Law 描述了模型性能如何随着模型参数规模、数据集大小和计算资源投入增长,展现出一定的幂律关系。讨论内容涵盖 Scaling Law 的核心公式、应用场景,潜在局限性及未来发展的可能性。时间轴00:00 - 开场与节目介绍01:02 - 什么是 Scaling Law?核心概念解析11:48 - Scaling Law 的现实意义及限制19:51 - 应用案例:OpenAI 的研究成果及行业观察23:16 - Scaling Law 的限制及未来发展方向24:16 - 结尾总结相关资源:OpenAI 的 Scaling Law 研究论文https://doi.org/10.48550/arXiv.2001.08361Hestness, Joel; Narang, Sharan; Ardalani, Newsha; Diamos, Gregory; Jun, Heewoo; Kianinejad, Hassan; Patwary, Md Mostofa Ali; Yang, Yang; Zhou, Yanqi (2017-12-01). "Deep Learning Scaling is Predictable, Empirically"https://doi.org/10.48550/arXiv.1712.00409Hoffmann, Jordan; Borgeaud, Sebastian; Mensch, Arthur; Buchatskaya, Elena; Cai, Trevor; Rutherford, Eliza; Casas, Diego de Las; Hendricks, Lisa Anne; Welbl, Johannes; Clark, Aidan; Hennigan, Tom; Noland, Eric; Millican, Katie; Driessche, George van den; Damoc, Bogdan (2022-03-29). "Training Compute-Optimal Large Language Models"https://doi.org/10.48550/arXiv.2203.15556联系我们如果喜欢本期节目,请订阅、分享,欢迎留下宝贵的意见,如有错误请多指正,新人上路,请多指教!联系方式:公众号【孙霄逸 Xiaoyi】特别鸣谢:本期主播:马克孙, 老柏感谢团队:对本期节目的策划与剪辑支持。

Type above to search every episode's transcript for a word or phrase. Matches are scoped to this podcast.

Searching…

We're indexing this podcast's transcripts for the first time — this can take a minute or two. We'll show results as soon as they're ready.

No matches for "" in this podcast's transcripts.

Showing of matches

No topics indexed yet for this podcast.

Loading reviews...

ABOUT THIS SHOW

脑袋尖尖,我们是伦敦的理工学生,想要做一个分享科技新闻,探索前沿科技和学术论文的播客。

HOSTED BY

马克孙

CATEGORIES

URL copied to clipboard!