arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.09820cs.LGstat.ML
收藏

GPEX:一个用于解释人工神经网络的框架

GPEX, A Framework For Interpreting Artificial Neural Networks

University of Alberta(阿尔伯塔大学)

Amir Akbarnejad, Gilbert Bigras, Nilanjan Ray

AI总结:

本文提出GPEX框架,通过证据下界使高斯过程匹配任意人工神经网络输出,并利用可扩展计算和核函数解释网络决策,在5个数据集上验证了其有效性。

摘要:

高斯过程(GPs)与深度人工神经网络(ANNs)之间的类比已引起广泛关注,并展现出揭开深度人工神经网络黑箱的潜力。现有理论工作对人工神经网络施加了严格假设(例如要求所有中间层都足够宽,或使用特定的激活函数)。在近期的深度架构中,满足这些理论假设十分困难,而且随着新的深度架构不断出现,这些理论条件也需要进一步细化。在本文中,我们推导出一个证据下界,该下界鼓励高斯过程的后验匹配人工神经网络的输出,而对人工神经网络没有任何要求。使用我们的方法,我们发现在5个数据集上,只有一部分理论假设是充分的。事实上,在我们的实验中,我们使用了普通的ResNet-18或前馈骨干网络,仅在末端设置了一个宽层。训练高斯过程的一个局限性在于其可扩展性受限于诱导点的数量。我们使用了新的计算技术,使我们能够训练具有数十万个诱导点并带有GPU加速的高斯过程。如我们的实验所示,这样做对于在5个数据集上实现高斯过程与人工神经网络之间的紧密匹配至关重要。我们将我们的方法实现为一个公开可用的工具,名为GPEX:https://github.com/amirakbarnejad/gpex。在5个数据集(4个图像数据集和1个生物数据集)以及具有2种功能类型(分类器或注意力机制)的人工神经网络上,我们能够找到输出与相应人工神经网络紧密匹配的高斯过程。在将高斯过程与人工神经网络匹配之后,我们使用高斯过程的核函数来解释人工神经网络的决策。我们提供了超过200个解释(论文中约30个解释,其余在补充材料中),这些解释对人类具有高度可解释性,并展示了所获得的高斯过程揭开人工神经网络决策的能力。

英文摘要:

The analogy between Gaussian processes (GPs) and deep artificial neural networks (ANNs) has received a lot of interest, and has shown promise to unbox the blackbox of deep ANNs. Existing theoretical works put strict assumptions on the ANN (e.g. requiring all intermediate layers to be wide, or using specific activation functions). Accommodating those theoretical assumptions is hard in recent deep architectures, and those theoretical conditions need refinement as new deep architectures emerge. In this paper we derive an evidence lower-bound that encourages the GP's posterior to match the ANN's output without any requirement on the ANN. Using our method we find out that on 5 datasets, only a subset of those theoretical assumptions are sufficient. Indeed, in our experiments we used a normal ResNet-18 or feed-forward backbone with a single wide layer in the end. One limitation of training GPs is the lack of scalability with respect to the number of inducing points. We use novel computational techniques that allow us to train GPs with hundreds of thousands of inducing points and with GPU acceleration. As shown in our experiments, doing so has been essential to get a close match between the GPs and the ANNs on 5 datasets. We implement our method as a publicly available tool called GPEX: https://github.com/amirakbarnejad/gpex. On 5 datasets (4 image datasets, and 1 biological dataset) and ANNs with 2 types of functionality (classifier or attention-mechanism) we were able to find GPs whose outputs closely match those of the corresponding ANNs. After matching the GPs to the ANNs, we used the GPs' kernel functions to explain the ANNs' decisions. We provide more than 200 explanations (around 30 explanations in the paper and the rest in the supplementary) which are highly interpretable by humans and show the ability of the obtained GPs to unbox the ANNs' decisions.

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。

↑