In this paper, we look at cross-domain few-shot classification which presents the challenging task of learning new classes in unseen domains with few labelled examples. Existing methods, though somewhat effective, encounter several limitations, which we address in this work through two significant improvements. First, to address overfitting associated with fine-tuning a large number of parameters on small datasets, we introduce a lightweight parameter-efficient adaptation strategy. This strategy employs a linear transformation of pre-trained features, significantly reducing the trainable parameter count. Second, we replace the traditional nearest centroid classifier with a variance-aware loss function, enhancing the model's sensitivity to the inter- and intra-class variances within the training set for improved clustering in feature space. Empirical evaluations on the Meta-Dataset benchmark showcase that our approach not only improves accuracy up to 7.7% and 5.3% on seen and unseen datasets respectively but also achieves this performance while being at least ~3x more parameter-efficient than existing methods, establishing a new state-of-the-art in cross-domain few-shot learning. Our code can be found at https://github.com/rashindrie/DIPA.

翻译：本文研究跨域小样本分类问题，该问题面临在未见过的领域中仅用少量标注样本学习新类别的挑战。现有方法虽有一定效果，但存在若干局限，本文通过两项重要改进加以解决。首先，针对小数据集微调大量参数易导致的过拟合问题，我们引入一种轻量级参数高效适配策略。该策略通过预训练特征的线性变换显著降低可训练参数量。其次，我们用方差感知损失函数替代传统最近质心分类器，增强模型对训练集中类间与类内方差的敏感度，从而优化特征空间中的聚类效果。在Meta-Dataset基准上的实验评估表明，我们的方法在已见和未见数据集上的准确率分别提升达7.7%和5.3%，同时参数量效率至少比现有方法高约3倍，确立了跨域小样本学习领域的新标杆。代码开源于https://github.com/rashindrie/DIPA。