arXiv subjects

Kayoung Ban

Publications and source records attributed to Kayoung Ban.

8 recordsLinked to original sources

arXiv2025

Diffolio: A Diffusion Model for Multivariate Probabilistic Financial Time-Series Forecasting and Portfolio Construction

Probabilistic forecasting is crucial in multivariate financial time-series for constructing efficient portfolios that account for complex cross-sectional dependencies. In this paper, we propose Diffolio, a diffusion model designed for multivariate financial time-series forecasting and portfolio construction. Diffolio employs a denoising network with a hierarchical attention architecture, comprising both asset-level and market-level layers. Furthermore, to better reflect cross-sectional correlations, we introduce a correlation-guided regularizer informed by a stable estimate of the target correlation matrix. This structure effectively extracts salient features not only from historical returns but also from asset-specific and systematic covariates, significantly enhancing the performance of forecasts and portfolios. Experimental results on the daily excess returns of 12 industry portfolios show that Diffolio outperforms various probabilistic forecasting baselines in multivariate forecasting accuracy and portfolio performance. Moreover, in portfolio experiments, portfolios constructed from Diffolio's forecasts show consistently robust performance, thereby outperforming those from benchmarks by achieving higher Sharpe ratios for the mean-variance tangency portfolio and higher certainty equivalents for the growth-optimal portfolio. These results demonstrate the superiority of our proposed Diffolio in terms of not only statistical accuracy but also economic significance.

cs.CE↗

arXiv2025

Quantum Integration Networks for Efficient Monte Carlo in High-Energy Physics

Monte Carlo methods play a central role in particle physics, where they are indispensable for simulating scattering processes, modeling detector responses, and performing multi-dimensional integrals. However, traditional Monte Carlo methods often suffer from slow convergence and insufficient precision, particularly for functions with singular features such as rapidly varying regions or narrow peaks. Quantum circuits provide a promising alternative: compared to conventional neural networks, they can achieve rich expressivity with fewer parameters, and the parameter-shift rule provides an exact analytic form for circuit gradients, ensuring precise optimization. Motivated by these advantages, we investigate how sampling strategies and loss functions affect integration efficiency within the \textbf{Quantum Integration Network} (QuInt-Net). We compare adaptive and non-adaptive sampling approaches and examine the impact of different loss functions on accuracy and convergence. Furthermore, we explore three quantum circuit architectures for numerical integration: the data re-uploading model, the quantum signal processing protocol, and deterministic quantum computation with one qubit. The results provide new insights into optimizing QuInt-Nets for applications in high energy physics.

quant-ph↗

arXiv2024

LeStrat-Net: Lebesgue style stratification for Monte Carlo simulations powered by machine learning

We develop a machine learning algorithm to turn around stratification in Monte Carlo sampling. We use a different way to divide the domain space of the integrand, based on the height of the function being sampled, similar to what is done in Lebesgue integration. This means that isocontours of the function define regions that can have any shape depending on the behavior of the function. We take advantage of the capacity of neural networks to learn complicated functions in order to predict these complicated divisions and preclassify large samples of the domain space. From this preclassification we can select the required number of points to perform a number of tasks such as variance reduction, integration and even event selection. The network ultimately defines the regions with what it learned and is also used to calculate the multi-dimensional volume of each region.

hep-ph↗

arXiv2023

Exploring the Synergy of Kinematics and Dynamics for Collider Physics

In collider experiments, an event is characterized by two distinct yet mutually complementary features: the `global features' and the `local features'. Kinematic information such as the event topology of a hard process, masses, and spins of particles comprises global features spanning the entire phase space. This global feature can be inferred from reconstructed objects. In contrast, representations of particles in gauge groups, such as Quantum Chromodynamics (QCD), offer localized features revealing the dynamics of an underlying theory. These local features, particularly observed in the patterns of radiation as raw data in various detector components, complement the global kinematic features. In this letter, we propose a simple but effective neural network architecture that seamlessly integrates information from both kinematics and QCD to enhance the signal sensitivity at colliders.

hep-ph↗

arXiv2022

DeeLeMa: Missing information search with Deep Learning for Mass estimation

We introduce DeeLeMa, a deep learning-based network for the analysis of energy and momentum in high-energy particle collisions. This novel approach is specifically designed to address the challenge of analyzing collision events with multiple invisible particles, which are prevalent in many high-energy physics experiments. DeeLeMa is constructed based on the kinematic constraints and symmetry of the event topologies. We show that DeeLeMa can robustly estimate mass distribution even in the presence of combinatorial uncertainties and detector smearing effects. The approach is flexible and can be applied to various event topologies by leveraging the relevant kinematic symmetries. This work opens up exciting opportunities for the analysis of high-energy particle collision data, and we believe that DeeLeMa has the potential to become a valuable tool for the high-energy physics community.

hep-ph↗

arXiv2022

Phenomenological implications on a hidden sector from the Festina Lente bound

We apply the Festina Lente (FL) bound on a hidden sector with $U(1)$ gauge symmetries. Since the FL bound puts a lower bound on masses of particles charged under the $U(1)$ gauge symmetries, it is possible to constrain the hidden sector even with a tiny coupling to the Standard Model. In particular, we focus on the phenomenological implications of the FL bound on milli-charged particles, which naturally arise when kinetic mixing between the photon and the hidden photon is allowed. It turns out that the milli-charged particle with the mass $M\lesssim 5$ meV is prohibited by the FL bound in the case of a single hidden $U(1)$, independent of the value of kinetic mixing. This bound is crucial when bosonic dark matter is taken in consideration in this framework: the fuzzy bosonic dark matter models requesting minuscule masses are ruled out by the FL bound if the longevity of dark matter is protected by the hidden gauge symmetry.

hep-ph↗

arXiv2021

A comprehensive study of vector leptoquark with $U(1)_{B_3-L_2}$ on the $B$-meson and Muon g-2 anomalies

Recently reported anomalies in various $B$ meson decays and also in the anomalous magnetic moment of muon $(g-2)_\mu$ motivate us to consider a particular extension of the standard model incorporating new interactions in lepton and quark sectors simultaneously. Our minimal choice would be leptoquark. In particular, we take vector leptoquark ($U_1$) and comprehensively study all related observables including ${(g-2)_{\mu}},\ R_{K^{(*)}},\ R_{D^{(*)}}$, $B \to (K) \ell \ell' $ where $\ell\ell'$ are various combinations of $\mu$ and $\tau$, and also lepton flavor violation in the $\tau$ decays. We find that a hybrid scenario with additional $U(1)_{B_3-L_2}$ gauge boson provides a common explanation of all these anomalies.

hep-ph↗

arXiv2020

Search for new light vector boson using $J/\Psi$ at BESIII and Belle II

We investigate various search strategies for light vector boson $X$ in $\mathcal{O}(10)~{\rm MeV}$ mass range using $J/\Psi$ associated channels at BESIII and Belle II: (i) $J/\Psi \to \eta_c X$ with $10^{10} J/\Psi$s at BESIII, (ii) $J/\Psi (\eta_c +X) +\ell \bar{\ell}$ production at Belle~II, and (iii) $J/\Psi +X$ with the displaced vertex in $X\to e^+e^-$ decay are analyzed and the future sensitivities at Belle II with 50 ${\rm ab}^{-1}$ luminosity are comprehensively studied. By requiring the displaced vertex to be within the beam pipe, the third method results in nearly background-free analysis, and the vector boson-electron coupling and the vector boson mass can be probed in the unprecedented range, $10^{-4}\leq |\varepsilon_e| \leq 10^{-3}$ and $9~{\rm MeV}\leq m_X\leq 100 {\rm MeV}$ with 50 ${\rm ab}^{-1}$ at Belle II. This covers the favored signal region of $^8{\rm Be}^*$ anomaly recently reported by Atomki experiment with $m_X \simeq 17~{\rm MeV}$.

hep-ph↗