arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.14780cs.LGcs.AI
收藏

GINN-KAN:可解释性流水线及其在物理信息神经网络中的应用

GINN-KAN: Interpretability pipelining with applications in Physics Informed Neural Networks

University of Melbourne(墨尔本大学)

Nisal Ranasinghe, Yu Xia, Sachith Seneviratne, Saman Halgamuge

AI总结:

本文提出可解释性流水线概念,融合GINN与KAN的优势构建了新型可解释神经网络GINN-KAN,在Feynman数据集及15个偏微分方程上表现出优于原模型和黑盒PINNs的求解能力。

摘要:

神经网络是强大的函数逼近器,然而其“黑盒”特性往往使其变得不透明且难以解释。尽管存在许多事后解释方法,但它们通常无法捕捉网络潜在的推理过程。一个真正可解释的神经网络应该像传统模型一样使用反向传播等技术进行训练,但还能为学习到的输入输出关系提供洞察。在这项工作中,我们引入了可解释性流水线的概念,以结合多种可解释性技术,从而超越任何单一技术的表现。为此,我们首先评估了几种有望实现此类可解释性的架构,特别关注了两个近期模型,选择它们是因为它们有望将可解释性纳入标准神经网络架构中,同时仍能利用反向传播:Growing Interpretable Neural Network (GINN) 和 Kolmogorov Arnold Networks (KAN)。我们分析了各自的局限性和优势,并引入了一种新颖的可解释神经网络 GINN-KAN,它综合了这两个模型的优点。在 Feynman 符号回归基准数据集上进行测试时,GINN-KAN 的表现优于 GINN 和 KAN。为了突出该方法的能力和泛化性,我们将 GINN-KAN 作为 Physics-Informed Neural Networks (PINNs) 中传统黑盒网络的替代方案。我们预计这将在深度学习流水线应用于自然科学方面产生深远影响。我们使用这种可解释的 PINN 在 15 个不同的偏微分方程上进行了实验,结果表明,增强 GINN-KAN 的 PINN 在求解微分方程方面优于使用黑盒网络的 PINN,并且超越了 GINN 和 KAN 的能力。

英文摘要:

Neural networks are powerful function approximators, yet their ``black-box" nature often renders them opaque and difficult to interpret. While many post-hoc explanation methods exist, they typically fail to capture the underlying reasoning processes of the networks. A truly interpretable neural network would be trained similarly to conventional models using techniques such as backpropagation, but additionally provide insights into the learned input-output relationships. In this work, we introduce the concept of interpretability pipelineing, to incorporate multiple interpretability techniques to outperform each individual technique. To this end, we first evaluate several architectures that promise such interpretability, with a particular focus on two recent models selected for their potential to incorporate interpretability into standard neural network architectures while still leveraging backpropagation: the Growing Interpretable Neural Network (GINN) and Kolmogorov Arnold Networks (KAN). We analyze the limitations and strengths of each and introduce a novel interpretable neural network GINN-KAN that synthesizes the advantages of both models. When tested on the Feynman symbolic regression benchmark datasets, GINN-KAN outperforms both GINN and KAN. To highlight the capabilities and the generalizability of this approach, we position GINN-KAN as an alternative to conventional black-box networks in Physics-Informed Neural Networks (PINNs). We expect this to have far-reaching implications in the application of deep learning pipelines in the natural sciences. Our experiments with this interpretable PINN on 15 different partial differential equations demonstrate that GINN-KAN augmented PINNs outperform PINNs with black-box networks in solving differential equations and surpass the capabilities of both GINN and KAN.

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。

↑