论文标题
加快步伐：通过合奏伪标记的快速而简单的域适应
Pick up the PACE: Fast and Simple Domain Adaptation via Ensemble Pseudo-Labeling
论文作者
论文摘要
近年来，域适应性（DA）近年来受到了深度学习研究人员的广泛关注，因为它具有通过标记的数据提高测试准确性的潜力。大多数最先进的DA算法都需要大量的高参数调整，并且由于所需的批量较大，因此在计算密集型上进行了密集。在这项工作中，我们提出了一种快速而简单的DA方法，该方法由三个阶段组成：（1）通过协方差匹配，（2）伪标记和（3）结合。我们称此方法$ \ textbf {pace} $，对于$ \ textbf {p} $ seudo-labels，$ \ textbf {a} $ textbf {c} $ ovariances的赌注，以及$ \ textbf {e textbf {e} $ nsembles。在现代预预式骨架合奏中提取的固定特征的顶部训练了速度。在大多数基准的适应任务上，PACE超过了$ \ textbf {5-10 \％} $的先前最先进的适应任务，而无需训练神经网络。与最先进的DA方法相比，PACE将培训时间和超参数调谐时间分别减少了$ 82 \％$和$ 97 \％$。代码在此处发布：https：//github.com/chris210634/pace-domain-audaptation
Domain Adaptation (DA) has received widespread attention from deep learning researchers in recent years because of its potential to improve test accuracy with out-of-distribution labeled data. Most state-of-the-art DA algorithms require an extensive amount of hyperparameter tuning and are computationally intensive due to the large batch sizes required. In this work, we propose a fast and simple DA method consisting of three stages: (1) domain alignment by covariance matching, (2) pseudo-labeling, and (3) ensembling. We call this method $\textbf{PACE}$, for $\textbf{P}$seudo-labels, $\textbf{A}$lignment of $\textbf{C}$ovariances, and $\textbf{E}$nsembles. PACE is trained on top of fixed features extracted from an ensemble of modern pretrained backbones. PACE exceeds previous state-of-the-art by $\textbf{5 - 10 \%}$ on most benchmark adaptation tasks without training a neural network. PACE reduces training time and hyperparameter tuning time by $82\%$ and $97\%$, respectively, when compared to state-of-the-art DA methods. Code is released here: https://github.com/Chris210634/PACE-Domain-Adaptation
