Safe Domain Adaptation for Physics: Overcoming Nuisances, Label Shifts, and Simulation Priors

物理学中的安全域适应：克服干扰因素、标签偏移和模拟先验

Abstract: Domain adaptation is widely used to make neural networks trained on simulations applicable to experimental data. Its premise is that the two domains differ only in nuisances, and that the quantity of interest is distributed identically in both. In physics neither assumption holds: simulations can be wrong about the physics, and the distribution of the target quantity - an energy spectrum, a redshift distribution - is often the measurement itself.

摘要： 域适应（Domain adaptation）被广泛用于使在模拟数据上训练的神经网络能够应用于实验数据。其前提是两个域仅在干扰因素（nuisances）上存在差异，且目标量在两者中的分布完全相同。然而在物理学中，这两个假设都不成立：模拟可能在物理规律上存在偏差，而目标量的分布（如能谱、红移分布）往往正是测量本身。

We study the consequences of such mismatches on a toy air-shower benchmark in which a detector-response nuisance, a physical simulation shift, and an energy-spectrum shift can be switched on separately or together. Standard adversarial adaptation handles the conditional shifts, but once the two spectra differ it aligns them, replacing an uncontrolled bias by one anchored on the simulation prior.

我们通过一个空气簇射（air-shower）基准测试研究了此类不匹配带来的后果，在该测试中，探测器响应干扰、物理模拟偏移和能谱偏移可以分别或同时开启。标准的对抗性适应可以处理条件偏移，但一旦两个谱存在差异，它就会强行对齐它们，从而用一个锚定在模拟先验上的偏差取代了原本不受控的偏差。

We present adaptive domain adaptation, which reweights the simulated events so as to focus domain adaptation on the genuine physical mismatch alone. Since the predicted spectrum depends on model training configuration, we provide a label-free model selection rule for selecting the near-the-best operation point.

我们提出了一种自适应域适应方法，通过对模拟事件进行重加权，使域适应过程仅聚焦于真实的物理不匹配。由于预测的谱取决于模型训练配置，我们提供了一种无需标签的模型选择规则，用于选择接近最优的操作点。