Bayesian Optimization for Enhanced Language Models: Optimizing Acquisition Functions

TL;DR

This paper introduces a bilevel Bayesian Optimization approach with mixed acquisition functions to improve fine-tuning of large language models, leading to better generalization and up to 2.7% performance gains.

Contribution

It proposes a novel bilevel BO method combining multiple acquisition functions for more effective hyperparameter tuning of language models.

Findings

Improved generalization on GLUE tasks using the proposed method.

Fine-tuning performance increased by up to 2.7%.

Combining EI and UCB enhances optimization stability.

Abstract

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%.