用双层贝叶斯优化融合不同采集函数，提升大模型微调效果
Bayesian Optimization for Enhanced Language Models: Optimizing Acquisition Functions
- 设计双层贝叶斯优化框架，内层最小化训练损失，外层优化验证指标
- 在GLUE任务上，使用EI与UCB组合使微调性能提升最高达2.7%
- 针对不同采集函数敏感性动态选择策略，适合大模型超参调优场景
随着语言模型架构的多样化，下游任务中的微调变得愈发重要，但合适的超参数难以确定。尽管贝叶斯优化（BO）已被用于超参数调优，现有方法普遍忽视了采集函数的选择对优化过程的关键影响——采集函数决定了探索与利用的平衡，不同函数对训练损失和验证性能的敏感度各异。本文提出{双层-贝叶斯优化-加权平均}（Bilevel-BO-SWA）方法，将EI和UCB等采集函数融合于嵌套优化循环中：内层优化训练损失，外层优化验证指标。在RoBERTA-base模型上对GLUE任务的实验表明，采用EI与UCB组合可显著提升泛化能力，微调性能最高提升2.7%。
原文摘要 · Abstract (English)
With the rise of different language model architecture, fine-tuning is becoming even more important for down stream tasks Model gets messy, finding proper hyperparameters for fine-tuning. Although BO has been tried for hyperparameter tuning, most of the existing methods are oblivious to the fact that BO relies on careful choices of acquisition functions, which are essential components of BO that guide how much to explore versus exploit during the optimization process; Different acquisition functions have different levels of sensitivity towards training loss and validation performance; existing methods often just apply an acquisition function no matter if the training and validation performance are sensitive to the acquisition function or not. This work introduces{Bilevel - BO - SWA}, a model fusion approach coupled with a bilevel BO strategy to improve the fine - tunning of large language models. Our work on mixture of acquisition functions like EI and UCB into nested opt loops, where inner loop perform minimization of training loss while outer loops optimized w.r.t. val metric. Experiments on GLUE tasks using RoBERTA - base show that when using EI and UCB, there is an improvement in generalization, and fine - tuning can be improved by up to 2.7%.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。