构建获胜团队：使用子模传递性估计方法选择源模型集成
Building a Winning Team: Selecting Source Model Ensembles using a Submodular Transferability Estimation Approach
AI总结：
提出OSBORN度量，基于最优传输和子模函数，综合领域差异、任务差异和模型内聚性，估计多源模型集成的可传递性，在图像分类和分割任务上优于现有方法。
摘要：
近年来，估计公开预训练模型对目标任务的可传递性在迁移学习任务中占据了重要地位。现有工作提出了各种度量标准，使用户无需逐一微调每个模型，即可从预训练模型池中选择一个模型并明确确定一个。随着可用预训练模型数量的增长和模型集成的普及，研究多个源模型对给定目标任务的可传递性也变得至关重要。少数现有研究在多源集成设置中仅使用分类层的输出研究可传递性，忽略了可能的领域或任务不匹配。此外，它们在选择源模型时忽略了最重要的因素，即模型之间的内聚性，这可能影响集成预测的性能和置信度。为解决这些不足，我们提出了一种新颖的基于最优传输的子模可传递性度量（OSBORN），用于估计模型集成对下游任务的可传递性。OSBORN综合考虑图像领域差异、任务差异和集成中模型的内聚性，提供可靠的可传递性估计。我们在图像分类和语义分割任务上评估了OSBORN的性能。我们的设置包括28个源数据集、11个目标数据集、5种模型架构和2种预训练方法。我们将我们的方法与当前最先进的度量标准MS-LEEP和E-LEEP进行基准比较，并一致地优于它们。
英文摘要：
Estimating the transferability of publicly available pretrained models to a target task has assumed an important place for transfer learning tasks in recent years. Existing efforts propose metrics that allow a user to choose one model from a pool of pre-trained models without having to fine-tune each model individually and identify one explicitly. With the growth in the number of available pre-trained models and the popularity of model ensembles, it also becomes essential to study the transferability of multiple-source models for a given target task. The few existing efforts study transferability in such multi-source ensemble settings using just the outputs of the classification layer and neglect possible domain or task mismatch. Moreover, they overlook the most important factor while selecting the source models, viz., the cohesiveness factor between them, which can impact the performance and confidence in the prediction of the ensemble. To address these gaps, we propose a novel Optimal tranSport-based suBmOdular tRaNsferability metric (OSBORN) to estimate the transferability of an ensemble of models to a downstream task. OSBORN collectively accounts for image domain difference, task difference, and cohesiveness of models in the ensemble to provide reliable estimates of transferability. We gauge the performance of OSBORN on both image classification and semantic segmentation tasks. Our setup includes 28 source datasets, 11 target datasets, 5 model architectures, and 2 pre-training methods. We benchmark our method against current state-of-the-art metrics MS-LEEP and E-LEEP, and outperform them consistently using the proposed approach.
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。
