arXiv subjects

Denny Lane B. Sombillo

Publications and source records attributed to Denny Lane B. Sombillo.

15 recordsLinked to original sources

arXiv2026

Revisiting the $Λ(1405)$ pole structure with convolutional neural networks

We revisit the long-standing ambiguity surrounding the complex pole structure of the $Λ(1405)$ resonance by reframing it as a classification problem for convolutional neural networks (CNNs). By training our models to recognize subtle geometric variations on nearly degenerate lineshapes using targeted differential feature on empirical CLAS data, we establish a data-driven consensus on large inference samples. Our results show that the analytic structure characterized by two poles on the $[bt]$ sheet and additional pole on the $[bb]$ sheet globally dominates. The inference-guided pole parameter extraction reveals that the $Λ(1405)$ is a two-state system, characterized by a molecular state sitting below the $\bar{K}N$ threshold and a non-molecular state lying above the $Σπ$ threshold.

hep-ph↗

arXiv2026

Deep learning topological inference-guided $T_{cc}^{+}$ pole parameter extraction

We perform a data-driven study of the doubly charmed tetraquark candidate $T_{cc}^+$. An ensemble of deep neural network classifiers, trained on synthetic amplitudes with controlled analytic structures, identifies a dominant pole topology characterized by an isolated pole on the $[bt]$ Riemann sheet which is robust against left-hand cut effects. A subsequent pole parameter extraction was performed via the uniformized $\mathcal{S}$-matrix and a complementary $\mathcal{K}$-matrix parameterization, which respectively provides a model-independent baseline and dynamical insight on the pole position and trajectory of the resonant state. Using this two-pronged approach, we submit that the $T_{cc}^{+}$ is a shallow $D^0D^{*+}$ bound state in the second Riemann sheet of the complex plane.

hep-ph↗

arXiv2025

Learning Pole Structures of Hadronic States using Predictive Uncertainty Estimation

Matching theoretical predictions to experimental data remains a central challenge in hadron spectroscopy. In particular, the identification of new hadronic states is difficult, as exotic signals near threshold can arise from a variety of physical mechanisms. A key diagnostic in this context is the pole structure of the scattering amplitude, but different configurations can produce similar signatures. The mapping between pole configurations and line shapes is especially ambiguous near the mass threshold, where analytic control is limited. In this work, we introduce an uncertainty-aware machine learning approach for classifying pole structures in $S$-matrix elements. Our method is based on an ensemble of classifier chains that provide both epistemic and aleatoric uncertainty estimates. We apply a rejection criterion based on predictive uncertainty, achieving a validation accuracy of nearly $95\%$ while discarding only a small fraction of high-uncertainty predictions. Trained on synthetic data with known pole structures, the model generalizes to previously unseen experimental data, including enhancements associated with the $P_{c\bar{c}}(4312)^+$ state observed by LHCb. In this, we infer a four-pole structure, representing the presence of a genuine compact pentaquark in the presence of a higher channel virtual state pole with non-vanishing width. While evaluated on this particular state, our framework is broadly applicable to other candidate hadronic states and offers a scalable tool for pole structure inference in scattering amplitudes.

hep-ph↗

arXiv2025

Line shape analysis of $Λ(1405)$ in $γp \rightarrow K^+Σ^-π^+$ reaction using convolutional neural network

Interpreting peaks or dips that appear in an invariant mass distribution is a recurring challenge in hadron physics. These enhancements can be ambiguous, especially near a two-hadron threshold since kinematical and dynamical effects play an important role in their nature. One such enhancement is an exotic baryon $Λ(1405)$ which was first observed in 1973. Despite the few available experimental data, the statistics of the measurements of $Λ(1405)$ have improved for line shape analysis. The present consensus is that it is a structure of two poles both on the second Riemann sheet. However, there are still investigations of other pole structures corresponding to $Λ(1405)$. Lately, the use of a deep neural network in analyzing these line shapes has been proven to be effective, especially in distinguishing pole structures. Thus, in this study, we develop a convolutional neural network, a type of DNN, to determine the general pole structure that corresponds to $Λ(1405)$ found in the $Σ^-π^+$ invariant mass distribution measured by CLAS in their experiment involving the $γp \rightarrow K^+Σπ$ reaction. The CNN is trained using a two-channel uniformized $S$-matrix allowing us to control the position and the corresponding Riemann sheet of the poles. Our preliminary results show that the trained CNN can accurately distinguish pole structures in the $Σ^-π^+$ invariant mass distribution and agrees with the present consensus of a two-pole structure. This supports the preceding works on the $Λ(1405)$ and requires a thorough analysis of $Σ^+π^-$ and $Σ^0π^0$ invariant mass spectra.

hep-ph↗

arXiv2025

Feature extraction in partial wave analysis using $K$-matrix approach

Structures in the invariant mass distribution are often linked to unstable intermediate states or resonances. In experiments, many signals are detected which have broad, overlapping or intricate profiles, which makes their characterization a formidable task. To ascertain whether or not these enhancements are resonances, and to determine their physical parameters, such as the resonance mass, coupling strength, resonance width, and quantum numbers, a tool known as partial wave analysis (PWA) is employed. To ensure unitarity, and to account for superposing states and non-resonant effects in modeling the scattering amplitude, $K$-matrix parametrization is utilized. To factor in the centrifugal effects on the decay rates arising from breakup processes in nonzero angular momenta, the Blatt-Weisskopf barrier factors are incorporated into the formulation. In this study, $K$-matrix parametrized differential cross sections were generated within near-threshold and resonance energy ranges, which served as the training and validation datasets for the Deep Neural Network (DNN). The Fully-Connected Neural Network (FCNN) architecture is applied to facilitate the classification task of this work, which is the discrimination of dominating partial waves in the scattering processes. The DNN model was designed to encompass various two-hadron scattering phenomena. As a stepping-off point, the values of the thresholds, coupling constants, resonance masses and widths, scattering energy ranges, and the decay channels used in this study found their physical inspiration from pion-nucleon ($π$$N$) scattering system in the $S$, $P$, $D$, $F$, and $G$ partial waves. The results of the model's performance showed that the DNN can distinguish the partial wave with resonances from the other partial waves with purely non-resonant contributions, at an accuracy of $69\%$.

hep-ph↗

arXiv2025

Effects of closely spaced thresholds on line shapes with near-threshold enhancement

Hidden-charm pentaquarks were first experimentally detected by LHCb in 2019, one of which is the exotic $P_{c\bar{c}}(4312)^{+}$ state. The nature of this state remains uncertain which may be attributed to the proximity of this observed enhancement to the meson-baryon thresholds. In this study, we introduce a coupled-channel approach using three-channel separable potential model to account for near-threshold effects on the observed signal. In particular, we assign $Σ_c^+\bar{D}^0$ and $Σ_c^{++}D^-$ as higher-mass channels for $P_{c \bar{c}}(4312)^+$ state. Moreover, this allows us to propose four pole configurations, where either a bound state or virtual state pole were placed near the higher-mass thresholds. Using this scheme, we study the near-threshold effects on the transition of different pole configurations across the unphysical Riemann sheets and examine their effects on the amplitude line shape. We found out that $P_{c \bar{c}}(4312)^+$ state may be interpreted as a virtual state below the $Σ_c^+\bar{D}^0$ which is consistent with our initial modeling assumptions. Our result also conforms with the previous analysis done by Joint Physics Analysis Center (JPAC) \cite{Fernandez-Ramirez:2019koa}, which leans toward similar interpretation. This may indicate that a model-dependent framework employing a relatively simple model, such as a separable potential, can somehow complement with some of the widely used parameterizations.

hep-ph↗

arXiv2024

A Deep Learning Framework for Disentangling Triangle Singularity and Pole-Based Enhancements

Enhancements in the invariant mass distribution or scattering cross-section are usually associated with resonances. However, the nature of exotic signals found near hadron-hadron thresholds remain a puzzle today due to the presence of experimental uncertainties. In fact, a purely kinematical triangle diagram is also capable of producing similar structures, but do not correspond to any unstable quantum state. In this paper, we report for the first time, that a deep neural network can be trained to distinguish triangle singularity from pole-based enhancements with a reasonably high accuracy of discrimination between the two seemingly identical line shapes. We also identify the type of triangle enhancement that can be misidentified as a dynamic pole structure. We apply our method to confirm that the $P_ψ^N(4312)^+$ state is not due to a triangle singularity, but is more consistent with a pole-based interpretation, as determined solely through pure line-shape analysis. Lastly, we explain how our method can be used as a model-selection framework useful in studying other exotic hadron candidates.

hep-ph↗

arXiv2024

Analysis of hidden-charm pentaquarks as triangle singularities via deep learning

Identifying the nature of near-threshold enhancements is hindered by the limited resolution of experimental data leading to multiple conflicting interpretations. A prominent example of ambiguous line shape is the set of pentaquark signals observed by LHCb in 2019. Some of these signals can be interpreted as hadronic molecule, compact state, virtual state, or due to a kinematical triangle mechanism. In this work, we leverage the model-selection capability of deep neural networks to analyze and identify the nature of $P_{c\bar{c}}(4457)^+$. We trained a set of deep neural networks using line shapes with enhancements produced by triangle singularities and those produced by nearby poles. The training dataset for the triangle enhancements are generated by using a set of hadrons satisfying the required mass condition. The training line shapes for the pole-based classifications are generated using uniformized independent $S$-matrix poles configured to appear close to the relevant threshold. We found that, despite the presence of experimental uncertainties, the triangle mechanism is ruled out by the experimental data. The results also suggest that the data favor the pole-shadow pair interpretation for the $P_{c\bar{c}}(4457)^+$, which corresponds to a characteristic pole structure involving a resonance in a two-channel scattering system. Our result is consistent with the initial analysis done by LHCb favoring the Breit-Wigner fit over the triangle singularity. The present analysis offers an alternative approach to studying line shapes, supplementing the standard fitting methods.

hep-ph↗

arXiv2024

Pole structure of $P_ψ^N(4312)^+$ via machine learning and uniformized S-matrix

We probed the pole structure of the $P_ψ^{N}(4312)^{+}$ using a trained deep neural network. The training dataset was generated using uniformized independent S-matrix poles to ensure that the obtained interpretation is as model-independent as possible. To prevent possible ambiguity in the interpretation of the pole structure, we included the contribution from the off-diagonal element of the S-matrix. Five out of the six neural networks we trained favor $P_ψ^{N}(4312)^{+}$ as possibly having a three-pole structure, with one pole on each of the unphysical sheets - a first in its report. The two poles can be associated to a pole-shadow pair which is a characteristic of a true resonance. On the other hand, the last pole is most likely associated with the coupled-channel effect. The combined effect of these poles produced a peak below the $Σ^{+}_C\bar{D}^0$ which mimic the line shape of a hadronic molecule.

hep-ph↗

arXiv2023

Interpretation of near-threshold peaks using the method of independent S-matrix poles

We propose a model-independent analysis of near-threshold enhancements using independent S-matrix poles. In this formulation, we constructed a Jost function with controllable zeros to ensure that no poles are generated on the physical Riemann sheet. We show that there is a possibility of misinterpreting the observed near-threshold signals if one utilized a limited parametrization and restrict the analysis to only one element of the S-matrix. Specifically, there is a possibility of the emergence of ambiguous pair of poles which are singularities of the full S-matrix but may not manifest in one of its elements. For a more concrete discussion, we focused on an effective two-channel scattering where the full S-matrix is a $2\times2$ matrix. We apply our method to the coupled two-channel analysis of the $P_ψ^N(4312)^+$ and found that the compact pentaquark interpretation cannot be ruled out yet.

hep-ph↗

arXiv2023

Formulation of causality-preserving quantum time of arrival theory

We revisit the quantum correction to the classical time of arrival to address the unphysical instantaneous arrival in the limit of zero initial momentum. In this study, we show that the vanishing of arrival time is due to the contamination of the causality-violating component of the initial wave packet. Motivated by this observation, we propose to update the temporal collapse mechanism in [Galapon E. A. 2009, Proc. R. Soc. A.46571-86] to incorporate the removal of causality-violating spectra of the arrival time operator. We found that the quantum correction to the classical arrival time is still observed. Thus, our analysis validates that the correction is an inherent consequence of quantizing a time observable and is not just some mathematical artifact of the theory. We also discuss the possible application of the theory in describing point interactions in particle physics and provide a possible explanation to the observed neutron's lifetime anomaly.

quant-ph↗

arXiv2021

Classifying near-threshold enhancement using deep neural network

One of the main issues in hadron spectroscopy is to identify the origin of threshold or near-threshold enhancement. Prior to our study, there is no straightforward way of distinguishing even the lowest channel threshold-enhancement of the nucleon-nucleon system using only the cross-sections. The difficulty lies in the proximity of either a bound or virtual state pole to the threshold which creates an almost identical structure in the scattering region. Identifying the nature of the pole causing the enhancement falls under the general classification problem and supervised machine learning using a feed-forward neural network is known to excel in this task. In this study, we discuss the basic idea behind deep neural network and how it can be used to identify the nature of the pole causing the enhancement. The applicability of the trained network can be explored by using an exact separable potential model to generate a validation dataset. We find that within some acceptable range of the cut-off parameter, the neural network gives high accuracy of inference. The result also reveals the important role played by the background singularities in the training dataset. Finally, we apply the method to nucleon-nucleon scattering data and show that the network was able to give the correct nature of pole, i.e. virtual pole for ${}^1S_0$ partial cross-section and bound state pole for ${}^3S_0$.

hep-ph↗

arXiv2021

Unveiling the pole structure of S-matrix using deep learning

Particle scattering is a powerful tool to unveil the nature of various subatomic phenomena. The key quantity is the scattering amplitude whose analytic structure carries the information of the quantum states. In this work, we demonstrate our first step attempt to extract the pole configuration of inelastic scatterings using the deep learning method. Among various problems, motivated by the recent new hadron phenomena, we develop a curriculum learning method of deep neural network to analyze coupled channel scattering problems. We show how effectively the method works to extract the pole configuration associated with resonances in the $πN$ scatterings.

hep-ph↗

arXiv2021

Model independent analysis of coupled-channel scattering: a deep learning approach

We develop a robust method to extract the pole configuration of a given partial-wave amplitude. In our approach, a deep neural network is constructed where the statistical errors of the experimental data are taken into account. The teaching dataset is constructed using a generic S-matrix parametrization, ensuring that all the poles produced are independent of each other. The inclusion of statistical error results into a noisy classification dataset which we should solve using the curriculum method. As an application, we use the elastic $πN$ amplitude in the $I(J^P)=1/2(1/2^{-})$ sector where $10^6$ amplitudes are produced by combining points in each error bar of the experimental data. We fed the amplitudes to the trained deep neural network and find that the enhancements in the $πN$ amplitude are caused by one pole in each nearby unphysical sheet and at most two poles in the distant sheet. Finally, we show that the extracted pole configurations are independent of the way points in each error bar are drawn and combined, demonstrating the statistical robustness of our method.

hep-ph↗

arXiv2020

Classifying Pole of Amplitude Using Deep Neural Network

Most of exotic resonances observed in the past decade appear as peak structure near some threshold. These near-threshold phenomena can be interpreted as genuine resonant states or enhanced threshold cusps. Apparently, there is no straightforward way of distinguishing the two structures. In this work, we employ the strength of deep feed-forward neural network in classifying objects with almost similar features. We construct a neural network model with scattering amplitude as input and nature of pole causing the enhancement as output. The training data is generated by an S-matrix satisfying the unitarity and analyticity requirements. Using the separable potential model, we generate a validation data set to measure the network's predictive power. We find that our trained neural network model gives high accuracy when the cut-off parameter of the validation data is within $400$-$800\mbox{ MeV}$. As a final test, we use the Nijmegen partial wave and potential models for nucleon-nucleon scattering and show that the network gives the correct nature of pole.

hep-ph↗