深度学习预测准确却难解释，揭示了算法与物理洞察的路径差异。
Is the end of Insight in Sight ?
- 用物理约束神经网络模拟稀薄气体，研究模型可解释性
- 训练后权重呈随机高斯分布，未体现物理规律痕迹
- 质疑可解释性是否应成为所有AI系统的通用标准
深度学习的兴起挑战了科学中长期存在的洞察理想——即通过发现底层机制来理解现象。在许多现代应用中，精确预测已无需可解释模型，引发对可解释性是否现实或有意义的争论。从物理学视角出发，我们通过一个具体案例进行探讨：训练一个物理信息神经网络（PINN）求解由玻尔兹曼方程支配的稀薄气体动力学问题。尽管系统结构清晰、基本定律明确，但训练后的网络权重表现为近似高斯分布的随机矩阵，未显示出任何物理原理的明显痕迹。这表明深度学习与传统模拟可能以截然不同的认知路径达到相同结果——前者基于统计插值，后者依赖机制洞察。这一发现对可解释人工智能的局限性提出关键质疑，并反思可解释性是否应作为人工智能推理的普遍标准。
原文摘要 · Abstract (English)
The rise of deep learning challenges the longstanding scientific ideal of insight - the human capacity to understand phenomena by uncovering underlying mechanisms. In many modern applications, accurate predictions no longer require interpretable models, prompting debate about whether explainability is a realistic or even meaningful goal. From our perspective in physics, we examine this tension through a concrete case study: a physics-informed neural network (PINN) trained on a rarefied gas dynamics problem governed by the Boltzmann equation. Despite the system's clear structure and well-understood governing laws, the trained network's weights resemble Gaussian-distributed random matrices, with no evident trace of the physical principles involved. This suggests that deep learning and traditional simulation may follow distinct cognitive paths to the same outcome - one grounded in mechanistic insight, the other in statistical interpolation. Our findings raise critical questions about the limits of explainable AI and whether interpretability can - or should-remain a universal standard in artificial reasoning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。