发表机构
National Institute of Technology, Rourkela; Birla Institute of Technology and Science - Pilani（印度国家理工学院鲁尔基分校; 比拉理工学院皮拉尼校区）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究利用机器学习对中子星观测数据分类，验证了能量-动量平方引力（EMSG）的印记，其中随机森林分类器准确率达99.85%，表明ML辅助观测可有效探测修正引力。
AI 中文摘要
中子星（NS）为在强场区域测试引力以及寻找广义相对论（GR）的偏差提供了独特的实验室。在这项工作中，我们研究了能量-动量平方引力（EMSG）对中子星结构的影响，并检验了是否可以利用监督机器学习（ML）从可观测的恒星属性中识别其特征。我们针对EMSG耦合参数α={-5.01,-2.50,0,+2.50,+5.01}×10^{-38} erg^{-1} cm^3，求解了约10^4个核物质状态方程（EOSs）的修正Tolman-Oppenheimer-Volkoff方程，并计算了每个恒星构型的引力质量M、半径R、无量纲潮汐形变率Λ以及基本f模振荡频率。在施加M、R和Λ的观测约束后，我们将数据集分为训练集和测试集，并采用随机森林（RF）、K近邻（KNN）、支持向量机（SVM）、逻辑回归（LR）和高斯朴素贝叶斯（GNB）来利用(M, R, Λ, f)对代表性扇区α={-5.01,0,+5.01}×10^{-38} erg^{-1} cm^3进行分类。RF分类器表现最佳，准确率约为99.85%，精确率、召回率和F1分数均超过99.8%，而KNN也实现了超过99%的准确率。几乎对角的混淆矩阵表明，与不同EMSG扇区相关的观测上可行的中子星构型在多维可观测空间中保持高度可分。我们的结果表明，即使在观测过滤之后，中子星可观测值仍保留EMSG的稳健特征，这使得机器学习辅助的中子星观测成为利用当前和未来多信使观测探测修正引力的有前景的补充方法。
英文摘要
Neutron stars (NSs) provide a unique laboratory for testing gravity in the strong-field regime and for searching for deviations from General Relativity (GR). In this work, we investigate the effects of Energy-Momentum Squared Gravity (EMSG) on NS structure and examine whether its signatures can be identified from observable stellar properties using supervised machine learning (ML). We solve the modified Tolman-Oppenheimer-Volkoff equations for approximately $10^{4}$ nuclear equations of state (EOSs) for EMSG coupling parameters $α=\{-5.01,-2.50,0,+2.50,+5.01\}\times10^{-38}\,\mathrm{erg}^{-1}\mathrm{cm}^{3}$, and calculate the gravitational mass $M$, radius $R$, dimensionless tidal deformability $Λ$, and fundamental $f$-mode oscillation frequency for each stellar configuration. Imposing observational constraints on $M$, $R$, and $Λ$, we split our datasets into train and test sets, and we employ Random Forest (RF), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression (LR), and Gaussian Naive Bayes (GNB) to classify the representative sectors $α=\{-5.01,0,+5.01\}\times10^{-38}\,\mathrm{erg}^{-1}\mathrm{cm}^{3}$ using $(M, R,Λ,f)$. The RF classifier performs best, achieving an accuracy of approximately $99.85\%$ with precision, recall, and F1-scores exceeding $99.8\%$, while KNN also achieves accuracy above $99\%$. The nearly diagonal confusion matrices demonstrate that the observationally viable NS configurations associated with different EMSG sectors remain highly separable in the multidimensional observable space. Our results show that NS observables retain robust signatures of EMSG even after observational filtering, establishing ML-assisted NS observations as a promising complementary approach for probing modified gravity with current and future multi-messenger observations.
CommentsThis is the first version of this work. Any comments, suggestions, or feedback would be greatly appreciated
