Ranking Prompt--Plateau Correlations in Gamma-Ray Bursts with Symbolic Regression
用符号回归对伽马射线暴中瞬时辐射-平台关联进行排序
Gamma-ray burst (GRB) correlations probe central engines, relativistic outflows, and radiation processes. The Dainotti relation is an anticorrelation between the X-ray plateau luminosity $L_a$ and rest-frame plateau duration $T_a^{\rm rest}$. We use symbolic regression to compare it with other low-dimensional correlations in 251 GRBs with Swift-XRT plateau properties and Swift-BAT prompt-emission parameters. We perform a Dainotti-guided search, in which $\log T_a^{\rm rest}$ is always present, and an unrestricted search over subsets of at most three variables. The guided search recovers the Dainotti dependence; a D'Agostini fit gives $\log L_a=(51.92\pm0.30)-(1.31\pm0.09)\log T_a^{\rm rest}$, with intrinsic scatter $σ_{\rm int}=0.82\pm0.05$ dex. In the unrestricted ranking of 377 feature subsets, the classical Dainotti relation remains competitive, within the top $\sim25\%$ of the tested search spaces. The highest-ranked screened expression is $\log L_a\approx0.855\log L_{\rm peak}-0.782\log T_a^{\rm rest}$, with RMSE $\approx0.46$ and $R^2\approx0.88$. The highest-ranked runs repeatedly reduce to this prompt--plateau structure, without retaining additional spectral, absorption, or afterglow-shape variables. A direct regression gives $\log L_a=4.718+0.776\log L_{\rm peak}-0.805\log T_a^{\rm rest}$, with $R_{\rm P}=0.91$ for 246 bursts. An errors-in-variables refit gives consistent coefficients. After removing shared luminosity-distance dependence, the relation retains non-negligible explanatory power ($R^2\approx0.62$), consistent with a possible intrinsic component. Restricted Monte Carlo null tests give $p\approx0.01$. Plateau luminosity therefore shows a coupled dependence on prompt luminosity and plateau duration.
展开 ▾首次用符号回归扫描377个特征子集,系统排序了Dainotti关系及其扩展;最佳关联仅包含瞬时峰值光度与平台时间,移除距离依赖后仍有显著残差,暗示内禀物理联系。
伽马射线暴平台阶段的经验关系始于 Dainotti 等人发现的平台光度与平台持续时间间的反相关(Dainotti+ 2008)。后续三维基本平面的研究(Dainotti+ 2016; Dainotti+ 2023)加入瞬时峰值光度,显著减小了内禀色散。本文利用符号回归(Cranmer 2023)对 251 个 GRB 的 377 个特征子集进行了无假设搜索与排序,结果显示经典 Dainotti 关系虽仍具竞争力(前约 25%),但最强的关联为 log L_a ≈ 0.855 log L_peak − 0.782 log T_a^rest,且额外的光谱、吸收或余辉形态变量未被保留,表明瞬时-平台联系可能由同一中心引擎驱动。该方法为时域天文学中自动发现和排序经验关联提供了一种可解释的路径,未来可推广至多信使数据。
预印本 2026-10-16 · 收录 2026-10-19