论文标题
二元图相似性：任务传输学习中初始化选择的通用框架
Duality Diagram Similarity: a generic framework for initialization selection in task transfer learning
论文作者
论文摘要
在本文中，我们解决了转移学习中的一个开放研究问题，鉴于几种预训练的模型，它正在选择模型初始化以在新任务上实现高性能。我们根据深神经网络（DNN）之间的二元图相似性（DDS）提出了一种新的高效和准确的方法。 DDS是代表和比较不同特征维度的数据的通用框架。我们通过测量17个任务任务和预测排名的实际转移学习绩效排名之间的对应关系来验证我们对任务数据集的方法。基于$ 17 \ TIMES17 $转移的基于DDS的排名需要少于2分钟，并且显示出高度相关性（0.86美元）与实际转移学习排名，在Taskomony Benchmark上以较大的保证金（$ 10 \％$）优于最先进的方法（$ 10 \％$ $）。我们还展示了我们针对新任务的模型选择方法的鲁棒性，即Pascal VOC语义分段。此外，我们证明我们的方法可以应用于在NYUV2和Pascal VOC数据集上的2D，3D和语义任务上转移DNN中的最佳图层位置。
In this paper, we tackle an open research question in transfer learning, which is selecting a model initialization to achieve high performance on a new task, given several pre-trained models. We propose a new highly efficient and accurate approach based on duality diagram similarity (DDS) between deep neural networks (DNNs). DDS is a generic framework to represent and compare data of different feature dimensions. We validate our approach on the Taskonomy dataset by measuring the correspondence between actual transfer learning performance rankings on 17 taskonomy tasks and predicted rankings. Computing DDS based ranking for $17\times17$ transfers requires less than 2 minutes and shows a high correlation ($0.86$) with actual transfer learning rankings, outperforming state-of-the-art methods by a large margin ($10\%$) on the Taskonomy benchmark. We also demonstrate the robustness of our model selection approach to a new task, namely Pascal VOC semantic segmentation. Additionally, we show that our method can be applied to select the best layer locations within a DNN for transfer learning on 2D, 3D and semantic tasks on NYUv2 and Pascal VOC datasets.
