arXiv subjects

Shi Pu

Publications and source records attributed to Shi Pu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Hydrodynamic effects on spin polarization along the beam direction in Au+Au and p+Pb collisions

We investigate hydrodynamic effects on the spin polarization of $Λ$ hyperons in Au+Au collisions at $\sqrt{s_{NN}} = 200$ GeV and p+Pb collisions at $\sqrt{s_{NN}} = 8.16$ TeV using the CLVisc hydrodynamic framework. We present numerical results for the second Fourier sine coefficient of the longitudinal spin polarization, $\langle P_{z} \sin 2(ϕ_{p} - Ψ_{2}) \rangle$, as a function of multiplicity (centrality) under three equilibrium scenarios: $Λ$ equilibrium, $s$-quark equilibrium, and isothermal equilibrium. We highlight the respective roles of thermal vorticity and the thermal-shear tensor in generating $\langle P_{z} \sin 2(ϕ_{p} - Ψ_{2}) \rangle$ across collision systems and scenarios.

nucl-th↗

arXiv2025

NCCR: to Evaluate the Robustness of Neural Networks and Adversarial Examples

Neural networks have received a lot of attention recently, and related security issues have come with it. Many studies have shown that neural networks are vulnerable to adversarial examples that have been artificially perturbed with modification, which is too small to be distinguishable by human perception. Different attacks and defenses have been proposed to solve these problems, but there is little research on evaluating the robustness of neural networks and their inputs. In this work, we propose a metric called the neuron cover change rate (NCCR) to measure the ability of deep learning models to resist attacks and the stability of adversarial examples. NCCR monitors alterations in the output of specifically chosen neurons when the input is perturbed, and networks with a smaller degree of variation are considered to be more robust. The results of the experiment on image recognition and the speaker recognition model show that our metrics can provide a good assessment of the robustness of neural networks or their inputs. It can also be used to detect whether an input is adversarial or not, as adversarial examples are always less robust.

cs.CR↗

arXiv2025

Distributed Learning over Arbitrary Topology: Linear Speed-Up with Polynomial Transient Time

We study a distributed learning problem in which $n$ agents, each with potentially heterogeneous local data, collaboratively minimize the sum of their local cost functions via peer-to-peer communication. We propose a novel algorithm, \emph{Spanning Tree Push-Pull} (STPP), which employs two spanning trees extracted from a general communication graph to distribute both model parameters and stochastic gradients. Unlike prior approaches that rely heavily on spectral gap properties, STPP leverages a more flexible topological characterization, enabling robust information flow and efficient updates. Theoretically, we prove that STPP achieves linear speedup and polynomial transient iteration complexity -- up to $\mathcal{O}(n^7)$ for smooth nonconvex objectives and $\tilde{\mathcal{O}}(n^3)$ for smooth strongly convex objectives -- under arbitrary network topologies. Moreover, compared with existing methods, STPP achieves faster convergence rates on sparse and non-regular topologies (e.g., directed rings) and reduces communication overhead on dense networks (e.g., static exponential graphs). Numerical experiments further demonstrate the strong performance of STPP across various graph architectures.

math.OC↗

arXiv2025

Is the shear induced spin polarization non-dissipative?

The shear-induced polarization plays a crucial role in understanding the local polarization of $Λ$ and $\overlineΛ$ hyperons. A key puzzle is whether the shear-induced polarization is non-dissipative or not. In this work, we analyzed the shear-induced polarization and the anomalous Hall effects using the entropy flow and H-theorem introduced from quantum (chiral) kinetic theory. While the shear-induced polarization and the anomalous Hall effect do not directly contribute to the entropy production rate, the perturbations associated with the shear tensor lead to an increase in entropy, similar to the role of the shear tensor in classical kinetic theory. We also examined these effects within the framework of linear response theory using Zubarev's approach. The analysis of time-reversal transformation on the spin current in coordinate space suggests that shear-induced polarization violates time-reversal symmetry and, therefore, should vanish. However, a similar analysis of the Wigner function in phase space does not impose any additional constraints on shear-induced polarization, allowing it to persist in phase space as expected. This discussion indicates that time-reversal analysis in coordinate space alone may not be sufficient to determine whether an effect is dissipative. Furthermore, our analysis based on Zubarev's approach suggests that these effects may indeed possess a dissipative nature. These findings highlight the limitations of the current theoretical framework in fully characterizing the dissipative properties of these phenomena.

hep-ph↗

arXiv2025

Relativistic spin hydrodynamics with antisymmetric spin tensors and an extension of the Bargmann-Michel-Telegdi equation

We derive a formulation of relativistic spin hydrodynamics with totally antisymmetric spin tensors that satisfy the Frenkel-Mathisson-Pirani condition. In our proposed spin hydrodynamics, the second law of thermodynamics is fulfilled by the spin-induced corrections in the heat flow, the viscous tensor, and the antisymmetric part of the energy-momentum tensor. These corrections are interpreted as the inverse spin Hall effect and the anomalous Hall effect in the nonrelativistic limit. We show that our evolution equation for the spin density is interpreted as an extension of the Bargmann-Michel-Telegdi equation known in relativistic many-body systems, including the Thomas precession term, the spin-rotation term, and new coupling terms between spin and hydrodynamic variables.

nucl-th↗

arXiv2025

Stochastic Push-Pull for Decentralized Nonconvex Optimization

To understand the convergence behavior of the Push-Pull method for decentralized optimization with stochastic gradients (Stochastic Push-Pull), this paper presents a comprehensive analysis. Specifically, we first clarify the algorithm's underlying assumptions, particularly those regarding the network structure and weight matrices. Then, to establish the convergence rate under smooth nonconvex objectives, we introduce a general analytical framework that not only encompasses a broad class of decentralized optimization algorithms, but also recovers or enhances several state-of-the-art results for distributed stochastic gradient tracking methods. A key highlight is the derivation of a sufficient condition under which the Stochastic Push-Pull algorithm achieves linear speedup, matching the scalability of centralized stochastic gradient methods -- a result not previously reported. Extensive numerical experiments validate our theoretical findings, demonstrating the algorithm's effectiveness and robustness across various decentralized optimization scenarios.

math.OC↗

arXiv2025

Asynchronous Decentralized SGD under Non-Convexity: A Block-Coordinate Descent Framework

Decentralized optimization has become vital for leveraging distributed data without central control, enhancing scalability and privacy. However, practical deployments face fundamental challenges due to heterogeneous computation speeds and unpredictable communication delays. This paper introduces a refined model of Asynchronous Decentralized Stochastic Gradient Descent (ADSGD) under practical assumptions of bounded computation and communication times. To understand the convergence of ADSGD, we first analyze Asynchronous Stochastic Block Coordinate Descent (ASBCD) as a tool, and then show that ADSGD converges under computation-delay-independent step sizes. The convergence result is established without assuming bounded data heterogeneity. Empirical experiments reveal that ADSGD outperforms existing methods in wall-clock convergence time across various scenarios. With its simplicity, efficiency in memory and communication, and resilience to communication and computation delays, ADSGD is well-suited for real-world decentralized learning tasks.

cs.LG↗

arXiv2025

Decentralized Min-Max Optimization with Gradient Tracking

This paper presents a novel distributed formulation of the min-max optimization problem. Such a formulation enables enhanced flexibility among agents when optimizing their maximization variables. To address the problem, we propose two distributed gradient methods over networks, termed Distributed Gradient Tracking Ascent (DGTA) and Distributed Stochastic Gradient Tracking Ascent (DSGTA). We demonstrate that DGTA achieves an iteration complexity of $\mathcal{O}(κ^2\varepsilon^{-2})$, and DSGTA attains a sample complexity of $\mathcal{O}(κ^3\varepsilon^{-4})$ for nonconvex strongly concave (NC-SC) objective functions. Both results match those of their centralized counterparts up to constant factors related to the communication network. Numerical experiments further demonstrate the superior empirical performance of the proposed algorithms compared to existing methods.

math.OC↗

arXiv2025

Nuclear deformation effects in photoproduction of $ρ$ mesons in ultraperipheral isobaric collisions

We have investigated the $ρ^{0}$ meson photoproduction in ultraperipheral isobaric collisions between $_{44}^{96}\textrm{Ru}+_{44}^{96}\textrm{Ru}$ and $_{40}^{96}\textrm{Zr}+_{40}^{96}\textrm{Zr}$ at $\sqrt{s_{NN}}=200$ GeV, employing the dipole model with the equivalent photon approximation. By implementing the Woods-Saxon distribution to represent the nuclear mass density, which is derived from density functional theory with an inclusion of nuclear deformation effects, we have calculated the transverse momentum $q_{T}$ spectra in isobaric collisions. We observe the characteristic dip behavior in these spectra, indicative of diffraction phenomena in high-energy physics. We notice that the deformation effects cause a nearly linear increase with $q_{T}^{2}$ for $q_{T}^{2}\lesssim0.015$ $\textrm{GeV}^{2}$, aligning with experimental observations. We offer a simple explanation for the observed behavior in these spectra by introducing the effective width of the nuclei in the thickness function. We also extend our discussion on the $ρ^{0}$ meson photoproduction with the targets $^{63}\textrm{Cu}$,$^{197}\textrm{Au}$, and $^{238}\textrm{U}$.

hep-ph↗

arXiv2025

An Accelerated Distributed Stochastic Gradient Method with Momentum

In this paper, we introduce an accelerated distributed stochastic gradient method with momentum for solving the distributed optimization problem, where a group of $n$ agents collaboratively minimize the average of the local objective functions over a connected network. The method, termed ``Distributed Stochastic Momentum Tracking (DSMT)'', is a single-loop algorithm that utilizes the momentum tracking technique as well as the Loopless Chebyshev Acceleration (LCA) method. We show that DSMT can asymptotically achieve comparable convergence rates as centralized stochastic gradient descent (SGD) method under a general variance condition regarding the stochastic gradients. Moreover, the number of iterations (transient times) required for DSMT to achieve such rates behaves as $\mathcal{O}(n^{5/3}/(1-λ))$ for minimizing general smooth objective functions, and $\mathcal{O}(\sqrt{n/(1-λ)})$ under the Polyak-Łojasiewicz (PL) condition. Here, the term $1-λ$ denotes the spectral gap of the mixing matrix related to the underlying network topology. Notably, the obtained results do not rely on multiple inter-node communications or stochastic gradient accumulation per iteration, and the transient times are the shortest under the setting to the best of our knowledge.

math.OC↗

arXiv2025

Radiative corrections on vortical spin polarization in hot QCD matter

We investigate the radiative corrections on spin polarization of relativistic fermions induced by vortical fields in thermal-equilibrium QCD matter at weak coupling. Such corrections stem from the self-energy gradients in quantum kinetic theory, which are further obtained by a more systematic and general approach through the Keldysh equation. By applying the hard-thermal-loop approximation, we obtain new corrections upon the spin-polarization spectrum and also the axial-charge current in connection to the axial/chiral vortical effect for massive quarks up to the leading order of the QCD coupling. Further influence on spin alignment of vector mesons from similar effects is also analyzed.

hep-ph↗

arXiv2025

Distributed Random Reshuffling Methods with Improved Convergence

This paper proposes two distributed random reshuffling methods, namely Gradient Tracking with Random Reshuffling (GT-RR) and Exact Diffusion with Random Reshuffling (ED-RR), to solve the distributed optimization problem over a connected network, where a set of agents aim to minimize the average of their local cost functions. Both algorithms invoke random reshuffling (RR) update for each agent, inherit favorable characteristics of RR for minimizing smooth nonconvex objective functions, and improve the performance of previous distributed random reshuffling methods both theoretically and empirically. Specifically, both GT-RR and ED-RR achieve the convergence rate of $O(1/[(1-λ)^{1/3}m^{1/3}T^{2/3}])$ in driving the (minimum) expected squared norm of the gradient to zero, where $T$ denotes the number of epochs, $m$ is the sample size for each agent, and $1-λ$ represents the spectral gap of the mixing matrix. When the objective functions further satisfy the Polyak-Łojasiewicz (PL) condition, we show GT-RR and ED-RR both achieve $O(1/[(1-λ)mT^2])$ convergence rate in terms of the averaged expected differences between the agents' function values and the global minimum value. Notably, both results are comparable to the convergence rates of centralized RR methods (up to constant factors depending on the network topology) and outperform those of previous distributed random reshuffling algorithms.

math.OC↗

arXiv2025

Collisional corrections to spin polarization from quantum kinetic theory using Chapman-Enskog expansion

We have investigated the collisional corrections to the spin polarization pseudo-vector, $δ\mathcal{P}^μ$, using quantum kinetic theory in Chapman-Enskog expansion. We derive the spin Boltzmann equation incorporating Møller scattering process. We further consider two distinct scenarios using hard thermal loop approximations for simplification. In scenario (I), the vector charge distribution function is treated as off-equilibrium under the validity domain of gradient expansion. Remarkably, the polarization induced by gradients of thermal chemical potential and shear viscous tensors are modified, but $δ\mathcal{P}_{\textrm{ }}^μ$ in this scenario does not depend on the coupling constant. In scenario (II), the vector charge distribution function is assumed to be in local thermal equilibrium. Then collisional corrections $δ\mathcal{P}_{\textrm{ }}^μ$ in this scenario are at $\mathcal{O}(\hbar^{2}\partial^{2})$. Additionally, we evaluate the $δ\mathcal{P}^μ$ using relaxation time approach for comparative analysis. Our results establish the theoretical framework necessary for the future numerical investigations on the interaction corrections to spin polarization.

hep-ph↗

arXiv2024

Spin alignment of vector mesons in local equilibrium by Zubarev's approach

We compute the $00$ element of the spin density matrix, denoted as $ρ_{00}$ and called the spin alignment, up to the second order of the gradient expansion in local equilibrium by Zubarev's approach. In the first order, we obtain $ρ_{00}=1/3$, meaning that the contributions from thermal vorticity and shear stress tensor are vanishing. The non-vanishing contributions to $ρ_{00}-1/3$ appear in the second order of gradients in the Belinfante and canonical cases. We also discuss the properties of the spin density matrix under the time reversal transformation. The effective transport coefficient for the spin alignment induced by the thermal shear stress tensor is T-odd in the first order, implying that the first order effect is dissipative.

hep-ph↗

arXiv2024

Linear Convergence Analysis of Single-loop Algorithm for Bilevel Optimization via Small-gain Theorem

Bilevel optimization has gained considerable attention due to its broad applicability across various fields. While several studies have investigated the convergence rates in the strongly-convex-strongly-convex (SC-SC) setting, no prior work has proven that a single-loop algorithm can achieve linear convergence. This paper employs a small-gain theorem in {robust control theory} to demonstrate that a single-loop algorithm based on the implicit function theorem attains a linear convergence rate of $\mathcal{O}(ρ^{k})$, where $ρ\in(0,1)$ is specified in Theorem 3. Specifically, We model the algorithm as a dynamical system by identifying its two interconnected components: the controller (the gradient or approximate gradient functions) and the plant (the update rule of variables). We prove that each component exhibits a bounded gain and that, with carefully designed step sizes, their cascade accommodates a product gain strictly less than one. Consequently, the overall algorithm can be proven to achieve a linear convergence rate, as guaranteed by the small-gain theorem. The gradient boundedness assumption adopted in the single-loop algorithm (\cite{hong2023two, chen2022single}) is replaced with a gradient Lipschitz assumption in Assumption 2.2. To the best of our knowledge, this work is first-known result on linear convergence for a single-loop algorithm.

math.OC↗

arXiv2024

A solvable model for spin polarizations with flow-momentum correspondence

We present an analytically solvable model based on the blast-wave picture of heavy-ion collisions with flow-momentum correspondence. It can describe the key features of spin polarizations in heavy-ion collisions. With the analytical solution, we can clearly show that the spin polarization with respect to the reaction plane is governed by the directed flow, while the spin polarization along the beam direction is governed by the ellipticity in flow and in transverse emission area. There is a symmetry between the contribution from the vorticity and from the shear stress tensor due to the flow-momentum correspondence. The solution can be improved systematically by perturbation method.

nucl-th↗

arXiv2024

B-ary Tree Push-Pull Method is Provably Efficient for Distributed Learning on Heterogeneous Data

This paper considers the distributed learning problem where a group of agents cooperatively minimizes the summation of their local cost functions based on peer-to-peer communication. Particularly, we propose a highly efficient algorithm, termed ``B-ary Tree Push-Pull'' (BTPP), that employs two B-ary spanning trees for distributing the information related to the parameters and stochastic gradients across the network. The simple method is efficient in communication since each agent interacts with at most $(B+1)$ neighbors per iteration. More importantly, BTPP achieves linear speedup for smooth nonconvex and strongly convex objective functions with only $\tilde{O}(n)$ and $\tilde{O}(1)$ transient iterations, respectively, significantly outperforming the state-of-the-art results to the best of our knowledge. Our code is available at https://github.com/ryou98/BTPP.

math.OC↗

arXiv2024

CEDAS: A Compressed Decentralized Stochastic Gradient Method with Improved Convergence

In this paper, we consider solving the distributed optimization problem over a multi-agent network under the communication restricted setting. We study a compressed decentralized stochastic gradient method, termed ``compressed exact diffusion with adaptive stepsizes (CEDAS)", and show the method asymptotically achieves comparable convergence rate as centralized { stochastic gradient descent (SGD)} for both smooth strongly convex objective functions and smooth nonconvex objective functions under unbiased compression operators. In particular, to our knowledge, CEDAS enjoys so far the shortest transient time (with respect to the graph specifics) for achieving the convergence rate of centralized SGD, which behaves as $\mathcal{O}(n{C^3}/(1-λ_2)^{2})$ under smooth strongly convex objective functions, and $\mathcal{O}(n^3{C^6}/(1-λ_2)^4)$ under smooth nonconvex objective functions, where $(1-λ_2)$ denotes the spectral gap of the mixing matrix, and $C>0$ is the compression-related parameter. In particular, CEDAS exhibits the shortest transient times when $C < \mathcal{O}(1/(1 - λ_2)^2)$, which is common in practice. Numerical experiments further demonstrate the effectiveness of the proposed algorithm.

math.OC↗