arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2403.10097cs.LGcs.AIcs.CV
收藏

自适应随机特征正则化在深度神经网络微调中的应用

Adaptive Random Feature Regularization on Fine-tuning Deep Neural Networks

NTT(日本电信电话株式会社);Kyoto University(京都大学)

Shin'ya Yamaguchi, Sekitoshi Kanai, Kazuki Adachi, Daiki Chijiwa

AI总结:

本文提出AdaRand,一种无需辅助源信息且计算成本合理的自适应随机特征正则化方法,通过最小化特征向量与随机参考向量的差距并动态更新条件分布,有效缓解微调过拟合,提升性能。

摘要:

虽然微调是训练深度神经网络的事实标准方法,但在使用小型目标数据集时,它仍然存在过拟合问题。以往的方法通过维持源数据集的知识或引入对比损失等正则化项来提高微调性能。然而,这些方法需要辅助源信息(如源标签或数据集)或大量额外计算。在本文中,我们提出了一种简单的方法,称为自适应随机特征正则化(AdaRand)。AdaRand帮助训练模型的特征提取器自适应地改变特征向量的分布,以适应下游分类任务,无需辅助源信息,且计算成本合理。为此,AdaRand最小化特征向量与从类条件高斯分布中采样的随机参考向量之间的差距。此外,AdaRand动态更新条件分布,以跟随当前更新的特征提取器,并平衡特征空间中类之间的距离。我们的实验表明,AdaRand优于其他需要辅助源信息和大量计算成本的微调正则化方法。

英文摘要:

While fine-tuning is a de facto standard method for training deep neural networks, it still suffers from overfitting when using small target datasets. Previous methods improve fine-tuning performance by maintaining knowledge of the source datasets or introducing regularization terms such as contrastive loss. However, these methods require auxiliary source information (e.g., source labels or datasets) or heavy additional computations. In this paper, we propose a simple method called adaptive random feature regularization (AdaRand). AdaRand helps the feature extractors of training models to adaptively change the distribution of feature vectors for downstream classification tasks without auxiliary source information and with reasonable computation costs. To this end, AdaRand minimizes the gap between feature vectors and random reference vectors that are sampled from class conditional Gaussian distributions. Furthermore, AdaRand dynamically updates the conditional distribution to follow the currently updated feature extractors and balance the distance between classes in feature spaces. Our experiments show that AdaRand outperforms the other fine-tuning regularization, which requires auxiliary source information and heavy computation costs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑