arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.09696cs.CV
收藏

用于课程学习的空间Transformer网络

Spatial Transformer Networks for Curriculum Learning

German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心(DFKI));TU Kaiserslautern(凯泽斯劳滕工业大学)

Fatemeh Azimi, Jean-Francois Jacques Nicolas Nies, Sebastian Palacio, Federico Raue, Jörn Hees, Andreas Dengel

摘要:

课程学习是一种生物启发的训练技术,其被广泛采用于机器学习,用于改进优化和关于收敛速率或获得的准确度的神经网络的更好训练。课程学习的主要概念是从较简单的任务开始训练,然后逐渐增加难度。因此,一个自然的问题是如何确定或生成这些更简单的任务。在这项工作中,我们从空间Transformer网络(STNs)的灵感,以形成一个简单的硬课程。由于STNs已被证明能够从输入图像中去除杂波,并在图像分类任务中获得更高的准确性,因此我们假设STNs处理的图像可以被视为更容易的任务,并用于课程学习。为此,我们研究了多种策略,用于塑造培训课程,使用STNs生成的数据。我们进行了各种实验上杂乱的MNIST和时尚MNIST数据集,在前者,我们得到了改善的$3.8$ PP的分类精度相比,基线。

英文摘要:

Curriculum learning is a bio-inspired training technique that is widely adopted to machine learning for improved optimization and better training of neural networks regarding the convergence rate or obtained accuracy. The main concept in curriculum learning is to start the training with simpler tasks and gradually increase the level of difficulty. Therefore, a natural question is how to determine or generate these simpler tasks. In this work, we take inspiration from Spatial Transformer Networks (STNs) in order to form an easy-to-hard curriculum. As STNs have been proven to be capable of removing the clutter from the input images and obtaining higher accuracy in image classification tasks, we hypothesize that images processed by STNs can be seen as easier tasks and utilized in the interest of curriculum learning. To this end, we study multiple strategies developed for shaping the training curriculum, using the data generated by STNs. We perform various experiments on cluttered MNIST and Fashion-MNIST datasets, where on the former, we obtain an improvement of $3.8$pp in classification accuracy compared to the baseline.

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。

↑