arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.11498cs.LGstat.ML
收藏

神经网络的相似性

Similarity of Neural Networks with Gradients

UC San Diego(加利福尼亚大学圣地亚哥分校);New York University(纽约大学);University of Warwick(华威大学);Amazon(亚马逊)

Shuai Tang, Wesley J. Maddox, Charlie Dickens, Tom Diethe, Andreas Damianou

摘要:

一个合适的相似性指标来比较学习过的神经网络在理解高度非线性函数的行为方面起着重要的作用,并且可以为进一步的理论分析和实证研究提供见解。在比较模型时,我们定义了两个关键步骤:首先,从学习模型中提取的表示,我们建议利用特征向量和梯度向量(在以前的工作中很大程度上被忽略)来设计神经网络的表示。其次,我们定义了所采用的相似性指数,它提供了所需的不变性,我们方便选择的草图技术,有效地比较各种数据集。从经验上讲,我们表明,所提出的方法提供了一个国家的最先进的方法来计算相似性的神经网络,独立训练不同的数据集和数据集定义的任务。

英文摘要:

A suitable similarity index for comparing learnt neural networks plays an important role in understanding the behaviour of the highly-nonlinear functions, and can provide insights on further theoretical analysis and empirical studies. We define two key steps when comparing models: firstly, the representation abstracted from the learnt model, where we propose to leverage both feature vectors and gradient ones (which are largely ignored in prior work) into designing the representation of a neural network. Secondly, we define the employed similarity index which gives desired invariance properties, and we facilitate the chosen ones with sketching techniques for comparing various datasets efficiently. Empirically, we show that the proposed approach provides a state-of-the-art method for computing similarity of neural networks that are trained independently on different datasets and the tasks defined by the datasets.

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。

↑