发表机构
Georgia Institute of Technology; Lawrence Berkeley National Laboratory; Max Planck Institute for Physics（佐治亚理工学院; 劳伦斯伯克利国家实验室; 马克斯·普朗克物理研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出两种混合推断技术，在接近NSBI性能的同时降低计算成本并保留可靠性，推荐潜在类别方法用于鲁棒高效推断。
AI 中文摘要
我们开发了两种混合技术，它们在接近神经模拟推断（NSBI）分析性能的同时，大幅降低了推断的计算成本，并保留了参数化方法的全部或部分可靠性保证。第一种方法具有广泛适用性，而第二种方法针对一类允许半参数NSBI表述的粒子物理分析量身定制。在原始灵敏度仅略有折损的情况下，这些方法代表了向离线分析中计算高效的NSBI迈出的重要一步，也为未来探索触发级应用打开了大门。基于我们的比较研究，我们推荐使用我们的第一种方法——潜在类别（Latent Categories），以实现鲁棒且高效的推断。
英文摘要
We develop two hybrid techniques that approach the performance of neural simulation-based inference (NSBI) analyses while substantially reducing the computational cost of inference and preserving some or all of the reliability guarantees of parametric methods. The first approach is broadly applicable, while the second is tailored to a class of particle physics analyses that admit a semi-parametric NSBI formulation. With only a modest compromise in raw sensitivity, these methods represent an important step toward computationally efficient NSBI in offline analyses and also open the door to the exploration of trigger-level applications in the future. Based on our comparison studies, we recommend the use of our first approach, Latent Categories, for robust and efficient inference.
Comments10 pages, 8 figures, code available at https://github.com/benevedes/hybrid-nsbi
