PAC-tuning:Fine-tuning Pretrained Language Models with PAC-driven Perturbed Gradient Descent
This addresses optimization and overfitting issues in fine-tuning for downstream tasks, particularly in few-shot scenarios, though it appears incremental as it builds on existing PAC-Bayes and gradient descent techniques.
The paper tackles the challenge of fine-tuning pretrained language models for few-shot learning by proposing PAC-tuning, a two-stage method that minimizes a PAC-Bayes generalization bound and uses perturbed gradient descent, achieving visible performance improvements over baselines on 5 GLUE benchmark tasks.
Fine-tuning pretrained language models (PLMs) for downstream tasks is a large-scale optimization problem, in which the choice of the training algorithm critically determines how well the trained model can generalize to unseen test data, especially in the context of few-shot learning. To achieve good generalization performance and avoid overfitting, techniques such as data augmentation and pruning are often applied. However, adding these regularizations necessitates heavy tuning of the hyperparameters of optimization algorithms, such as the popular Adam optimizer. In this paper, we propose a two-stage fine-tuning method, PAC-tuning, to address this optimization challenge. First, based on PAC-Bayes training, PAC-tuning directly minimizes the PAC-Bayes generalization bound to learn proper parameter distribution. Second, PAC-tuning modifies the gradient by injecting noise with the variance learned in the first stage into the model parameters during training, resulting in a variant of perturbed gradient descent (PGD). In the past, the few-shot scenario posed difficulties for PAC-Bayes training because the PAC-Bayes bound, when applied to large models with limited training data, might not be stringent. Our experimental results across 5 GLUE benchmark tasks demonstrate that PAC-tuning successfully handles the challenges of fine-tuning tasks and outperforms strong baseline methods by a visible margin, further confirming the potential to apply PAC training for any other settings where the Adam optimizer is currently used for training.