arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1911.03332cs.NEcs.LG
收藏

用于迁移学习的CNN特征鉴别度估计

Feature discriminativity estimation in CNNs for transfer learning

Barcelona Supercomputing Center(巴塞罗那超级计算中心);Universitat Politècnica de Catalunya - BarcelonaTech(加泰罗尼亚理工大学-巴塞罗那理工)

Victor Gimenez-Abalos, Armand Vilalta, Dario Garcia-Gasulla, Jesus Labarta, Eduard Ayguadé

摘要:

在卷积神经网络上进行特征提取的目的是重用为预训练模型学习的深度表示,以解决新的、可能不相关的问题。然而,考虑到这些网络的巨大规模,从所有层进行原始特征提取是不可行的。最近,提出了一种使用复杂度降低的监督方法,从而显著提高了迁移学习任务的性能。这种方法首先计算特征的鉴别力,然后使用为任务计算的阈值将它们离散化。在本文中,我们分析了这些阈值的行为,目的是找到一种方法来估计。经过综合研究,我们发现问题大小和阈值之间有很强的相关性,决定系数在90%以上。这些结果使我们能够提出一个统一的阈值估计模型,具有潜在的应用迁移学习任务。

英文摘要:

The purpose of feature extraction on convolutional neural networks is to reuse deep representations learnt for a pre-trained model to solve a new, potentially unrelated problem. However, raw feature extraction from all layers is unfeasible given the massive size of these networks. Recently, a supervised method using complexity reduction was proposed, resulting in significant improvements in performance for transfer learning tasks. This approach first computes the discriminative power of features, and then discretises them using thresholds computed for the task. In this paper, we analyse the behaviour of these thresholds, with the purpose of finding a methodology for their estimation. After a comprehensive study, we find a very strong correlation between problem size and threshold value, with coefficient of determination above 90%. These results allow us to propose a unified model for threshold estimation, with potential application to transfer learning tasks.

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。

↑