hiw3c 日报精选
arXiv cs.AI/CL/LG 网页快照

基于装袋方法的最优自回归思维链学习

原文标题 · Optimal Autoregressive Chain-of-Thought Learning via Bagging

作者: Awni Altabaa, John Lafferty, Omar Montasser | 我们研究如何从类别$\mathcal{F}$中学习一个自回归下一个标记生成器$f\colonΣ^\star\toΣ$,给定由$f$生成的长度为$T$的思维链(CoT)演示。我们提出一种基于装袋方法的简单算法,该算法输出经过训练以在训练数据的自助引导样本上符合CoT一致性的生成器的多数投票结果。对于二

来源:arXiv cs.AI/CL/LG · arxiv.org

正文为英文,可一键机器翻译(仅首次需要等待)

Computer Science > Machine Learning

Title: Optimal Autoregressive Chain-of-Thought Learning via Bagging

Abstract: We study how to learn an autoregressive next-token generator $f\colon\Sigma^\star\to\Sigma$ from a class $\mathcal{F}$, given chain-of-thought (CoT) demonstrations of length $T$ generated by $f$. We present a simple algorithm based on bagging, which outputs a majority vote of generators that are trained to be CoT-consistent on bootstrap samples of the training data. For binary token sets, it achieves the optimal sample complexity of $O((d_{\mathrm{VC}}+\log(1/\delta))/\epsilon)$, independent of the CoT length $T$, where $d_{\mathrm{VC}}$ is the VC dimension of $\mathcal{F}$. For finite token sets $\Sigma$, the VC dimension is replaced by a multiclass analogue, the Graph dimension $d_{\mathrm{G}}$, and the sample complexity is optimal up to a $\log|\Sigma|$ factor. The number of generators in the majority-vote ensemble is $O(\log(T/\delta))$, and we show that the ensemble can be distilled into a single generator in $\mathcal{F}$ at the expense of $n\log(nT)$ unlabeled prompts.

Submission history

Access Paper:

Additional Features

Current browse context:

References & Citations

BibTeX formatted citation

Bookmark

Bibliographic and Citation Tools

Code, Data and Media Associated with this Article

Demos

Recommenders and Search Tools

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .