arXiv subjects

Yongjoo Baek

Publications and source records attributed to Yongjoo Baek.

At least 19 recordsLinked to original sources

arXiv2026

Mass-induced Mpemba effect in a polymer-bead system

We propose a physically motivated model in which the Mpemba effect is induced by the system's inertia. The model describes a polymer undergoing a denaturation transition whose force-extension curve contains a weak-force plateau that slows relaxation. In the presence of inertia, a farther initial state can reach and cross the plateau sooner, producing the Mpemba effect. Increasing the bead mass broadens the range of initial conditions over which this mechanism operates. A similar mechanism also generates the inverse Mpemba effect.

cond-mat.soft↗

arXiv2026

Quantifying information flow along a stochastic trajectory

Stochastic information flow (SIF) quantifies information flow at the trajectory level, overcoming the limitations of conventional symmetric, ensemble-averaged measures. However, computational difficulties have hindered the empirical application of the SIF. In this work, we propose a scalable deep-learning method for estimating the SIF from general time-series data. Its applications to an exactly solvable two-particle model, Kuramoto oscillators, and empirical trajectories of interacting motile cells demonstrate the utility of SIF as a data-driven indicator of cooperative structures.

cond-mat.stat-mech↗

arXiv2026

Boltzmann Sampling by Diabatic Quantum Annealing

Boltzmann sampling is a central component of many computational frameworks, including numerous algorithms in machine learning. Although quantum annealers have been investigated as potential fast Boltzmann samplers, their dependence on environmental noise makes precise control of the effective temperature difficult, introducing uncertainty into the sampling process. As an alternative, we propose diabatic quantum annealing -- a faster, purely unitary process -- as a controllable Boltzmann sampler in which the effective temperature is determined by the annealing rate. Using the ferromagnetic Ising model and the Sherrington--Kirkpatrick model as test cases, we demonstrate that this method achieves rapid and accurate sampling in the high-temperature regime.

cond-mat.stat-mech↗

arXiv2025

Phase separation of chemokinetic active particles

Motility-induced phase separation (MIPS) is a well-studied nonequilibrium collective phenomenon observed in active particles. Recently, there has been growing interest in how coupling the self-propulsion of active particles to chemical degrees of freedom affects MIPS. Although the effects of chemotaxis on MIPS have been extensively studied, little is known about how chemokinesis affects MIPS. In this study, we demonstrate that various patterns can be induced when active particles consume chemicals and exhibit chemokinesis, where higher chemical concentrations enhance self-propulsion without causing alignment with the chemical gradient. We discover that MIPS is intensified if chemical consumption is proportional to particle density (as in the basal metabolic regime), but it is suppressed if chemical consumption is closely tied to particle motion (as in the active metabolic regime). While the former produces large-scale phase separation via coarsening, the latter suppresses the coarsening process, leading to microphase separation and oscillating patterns. We also derive a hydrodynamic theory that describes these findings.

cond-mat.soft↗

arXiv2025

Symmetry-breaking motility of an active hinge in a crowded channel

A recent experiment [Son et al., Soft Matter, 2024, 20,2777-2788] showed that self-propelled particles confined within a circular boundary filled with granular medium spontaneously form a motile cluster that stays on the boundary. This cluster exhibits persistent (counter)clockwise motion driven by symmetry breaking, which arises from a positive feedback between the asymmetry of the cluster and those of the surrounding granular medium. To investigate this symmetry-breaking mechanism in broader contexts, we propose and analyze the dynamics of an active hinge moving through a crowded two-dimensional channel. Through extensive numerical simulations, we find that the lifetime of the hinge's motile state varies nonmonotonically with both the packing fraction of the granular medium and the strength of self-propulsion. Furthermore, we observe an abrupt transition in the configuration of passive particles that sustain hinge motility as the hinge's maximum angle relative to the channel wall increases. These findings point to the possibility of designing superstructures composed of passive granular media doped with a small number of active elements, whose dynamics modes can be switched by tuning the properties of their components.

cond-mat.soft↗

arXiv2025

Thermodynamically consistent lattice Monte Carlo method for active particles

Recent years have seen a growing interest in the thermodynamic cost of dissipative structures formed by active particles. Given the strong finite-size effects of such systems, it is essential to develop efficient numerical approaches that discretize both space and time while preserving the original dynamics and thermodynamics of active particles in the continuum limit. To address this challenge, we propose two thermodynamically consistent kinetic Monte Carlo methods for active lattice gases, both of which correctly reproduce the continuum dynamics. One method follows the conventional Kawasaki dynamics, while the other incorporates an extra state-dependent prefactor in the transition rate to more accurately capture the self-propulsion velocity. We find that the error scales linearly with time step size and that the state-dependent prefactor improves accuracy at high Péclet numbers by a factor of $\mathrm{Pe}^2$. Our results are supported by rigorous proof of convergence as well as extensive simulations.

cond-mat.soft↗

arXiv2024

Exploring how deep learning decodes anomalous diffusion via Grad-CAM

While deep learning has been successfully applied to the data-driven classification of anomalous diffusion mechanisms, how the algorithm achieves the feat still remains a mystery. In this study, we use a well-known technique aimed at achieving explainable AI, namely the Gradient-weighted Class Activation Map (Grad-CAM), to investigate how deep learning (implemented by ResNets) recognizes the distinctive features of a particular anomalous diffusion model from the raw trajectory data. Our results show that Grad-CAM reveals the portions of the trajectory that hold crucial information about the underlying mechanism of anomalous diffusion, which can be utilized to enhance the robustness of the trained classifier against the measurement noise. Moreover, we observe that deep learning distills unique statistical characteristics of different diffusion mechanisms at various spatiotemporal scales, with larger-scale (smaller-scale) features identified at higher (lower) layers.

cs.LG↗

arXiv2024

Unified Hierarchical Relationship Between Thermodynamic Tradeoff Relations

Recent years have witnessed a surge of discoveries in the studies of thermodynamic inequalities: the thermodynamic uncertainty relation (TUR) and the entropic bound (EB) provide a lower bound on the entropy production (EP) in terms of nonequilibrium currents; the classical speed limit (CSL) expresses the lower bound on the EP using the geometry of probability distributions; the power-efficiency (PE) tradeoff dictates the maximum power achievable for a heat engine given the level of its thermal efficiency. In this study, we show that there exists a unified hierarchical structure encompassing all of these bounds, with the fundamental inequality given by a novel extension of the TUR (XTUR) that incorporates the most general range of current-like and state-dependent observables. By selecting more specific observables, the TUR and the EB follow from the XTUR, and the CSL and the PE tradeoff follow from the EB. Our derivations cover both Langevin and Markov jump systems, with the first proof of the EB for the Markov jump systems and a more generalized form of the CSL. We also present concrete examples of the EB for the Markov jump systems and the generalized CSL.

cond-mat.stat-mech↗

arXiv2024

$α$-divergence Improves the Entropy Production Estimation via Machine Learning

Recent years have seen a surge of interest in the algorithmic estimation of stochastic entropy production (EP) from trajectory data via machine learning. A crucial element of such algorithms is the identification of a loss function whose minimization guarantees the accurate EP estimation. In this study, we show that there exists a host of loss functions, namely those implementing a variational representation of the $α$-divergence, which can be used for the EP estimation. By fixing $α$ to a value between $-1$ and $0$, the $α$-NEEP (Neural Estimator for Entropy Production) exhibits a much more robust performance against strong nonequilibrium driving or slow dynamics, which adversely affects the existing method based on the Kullback-Leibler divergence ($α= 0$). In particular, the choice of $α= -0.5$ tends to yield the optimal results. To corroborate our findings, we present an exactly solvable simplification of the EP estimation problem, whose loss function landscape and stochastic properties give deeper intuition into the robustness of the $α$-NEEP.

cond-mat.stat-mech↗

arXiv2024

Symmetry-breaking motility of penetrable objects in active fluids

We investigate how a symmetric penetrable object immersed in an active fluid becomes motile due to a negative drag acting in the direction of its velocity. While similar phenomena have been reported only for active fluids that posses polar or nematic order, we demonstrate that such motility can occur even in active fluids without any preexisting order. The emergence of object motility is characterized by both continuous and discontinuous transitions associated with the symmetry-breaking bifurcation of the object's steady-state velocity. Furthermore, we also discuss the relevance of the transitions to the nonmonotonic particle-size dependence of the object's diffusion coefficient.

cond-mat.soft↗

arXiv2023

Tradeoff of generalization error in unsupervised learning

Finding the optimal model complexity that minimizes the generalization error (GE) is a key issue of machine learning. For the conventional supervised learning, this task typically involves the bias-variance tradeoff: lowering the bias by making the model more complex entails an increase in the variance. Meanwhile, little has been studied about whether the same tradeoff exists for unsupervised learning. In this study, we propose that unsupervised learning generally exhibits a two-component tradeoff of the GE, namely the model error and the data error -- using a more complex model reduces the model error at the cost of the data error, with the data error playing a more significant role for a smaller training dataset. This is corroborated by training the restricted Boltzmann machine to generate the configurations of the two-dimensional Ising model at a given temperature and the totally asymmetric simple exclusion process with given entry and exit rates. Our results also indicate that the optimal model tends to be more complex when the data to be learned are more complex.

cond-mat.stat-mech↗

arXiv2023

Effects of the self-propulsion parity on the efficiency of a fuel-consuming active heat engine

We propose a thermodynamically consistent, analytically tractable model of steady-state active heat engines driven by both temperature difference and a constant chemical driving. While the engine follows the dynamics of the Active Ornstein-Uhlenbeck Particle, its self-propulsion stems from the mechanochemical coupling with the fuel consumption dynamics, allowing for both even- and odd-parity self-propulsion forces. Using the standard methods of stochastic thermodynamics, we show that the entropy production of the engine satisfies the conventional Clausius relation, based on which we define the efficiency of the model that is bounded from above by the second law of thermodynamics. Using this framework, we obtain exact expressions for the efficiency at maximum power. The results show that the engine performance has a nonmonotonic dependence on the magnitude of the chemical driving, and that the even-parity (odd-parity) engines perform better when the size of the engine is smaller (larger) than the persistence length of the active particle. We also discuss the existence of a tighter upper bound on the efficiency of the odd-parity engines stemming from the detailed structure of the entropy production.

cond-mat.stat-mech↗

arXiv2020

Bodies in an Interacting Active Fluid: Far-Field Influence of a Single Body and Interaction Between Two Bodies

Because active particles break time-reversal symmetry, an active fluid can sustain currents even without an external drive. We show that when a passive body is placed in a fluid of pairwise interacting active particles, it generates long-range currents, corresponding to density and pressure gradients. By using a multipole expansion and a far-field constitutive relation, we show that the leading-order behavior of all three corresponds to a source dipole. Then, when two bodies or more are placed in the active fluid, generic long-range interactions between the bodies occur. We find these to be qualitatively different from other fluid mediated interactions, such as hydrodynamic or thermal Casimir. The interactions can be predicted by measuring a few single-body properties in separate experiments. Moreover, they are anisotropic and do not satisfy an action-reaction principle. These results extend previous results on non-interacting active particles. Our framework may point to a path towards self-assembly.

cond-mat.stat-mech↗

arXiv2019

Finite-size and finite-time effects in large deviation functions near dynamical symmetry breaking transitions

We introduce and study a class of particle hopping models consisting of a single box coupled to a pair of reservoirs. Despite being zero-dimensional, in the limit of large particle number and long observation time, the current and activity large deviation functions of the models can exhibit symmetry-breaking dynamical phase transitions. We characterize exactly the critical properties of these transitions, showing them to be direct analogues of previously studied phase transitions in extended systems. The simplicity of the model allows us to study features of dynamical phase transitions which are not readily accessible for extended systems. In particular, we quantify finite-size and finite-time scaling exponents using both numerical and theoretical arguments. Importantly, we identify an analogue of critical slowing near symmetry breaking transitions and suggest how this can be used in the numerical studies of large deviations. All of our results are also expected to hold for extended systems.

cond-mat.stat-mech↗

arXiv2019

Role of hubs in the synergistic spread of behavior

The spread of behavior in a society has two major features: the synergy of multiple spreaders and the dominance of hubs. While strong synergy is known to induce mixed-order transitions (MOTs) at percolation, the effects of hubs on the phenomena are yet to be clarified. By analytically solving the generalized epidemic process on random scale-free networks with the power-law degree distribution $p_k \sim k^{-α}$, we clarify how the dominance of hubs in social networks affects the conditions for MOTs. Our results show that, for $α< 4$, an abundance of hubs drive MOTs, even if a synergistic spreading event requires an arbitrarily large number of adjacent spreaders. In particular, for $2 < α< 3$, we find that a global cascade is possible even when only synergistic spreading events are allowed. These transition properties are substantially different from those of cooperative contagions, which are another class of synergistic cascading processes exhibiting MOTs.

physics.soc-ph↗

arXiv2017

Generic long-range interactions between passive bodies in an active fluid

Because active particles break time-reversal symmetry, a single non-spherical body placed in an active fluid generates currents. We show that when two or more passive bodies are placed in an active fluid these currents lead to long-range interactions. Using a multipole expansion we characterize their leading-order behaviors in terms of single-body properties and show that they decay as a power law with the distance between the bodies, are anisotropic, and do not obey an action--reaction principle. The interactions lead to rich dynamics of the bodies, illustrated by the spontaneous synchronized rotation of pinned non-chiral bodies and the formation of traveling bound pairs. The occurrence of these phenomena depends on tunable properties of the bodies, thus opening new possibilities for self-assembly mediated by active fluids.

cond-mat.stat-mech↗

arXiv2017

Dynamical phase transitions in the current distribution of driven diffusive channels

We study singularities in the large deviation function of the time-averaged current of diffusive systems connected to two reservoirs. A set of conditions for the occurrence of phase transitions, both first and second order, are obtained by deriving Landau theories. First-order transitions occur in the absence of a particle-hole symmetry, while second-order occur in its presence and are associated with a symmetry breaking. The analysis is done in two distinct statistical ensembles, shedding light on previous results. In addition, we also provide an exact solution of a model exhibiting a second-order symmetry-breaking transition.

cond-mat.stat-mech↗

arXiv2017

Effects of a local defect on one-dimensional nonlinear surface growth

The slow-bond problem is a long-standing question about the minimal strength $ε_\mathrm{c}$ of a local defect with global effects on the Kardar--Parisi--Zhang (KPZ) universality class. A consensus on the issue has been delayed due to the discrepancy between various analytical predictions claiming $ε_\mathrm{c} = 0$ and numerical observations claiming $ε_\mathrm{c} > 0$. We revisit the problem via finite-size scaling analyses of the slow-bond effects, which are tested for different boundary conditions through extensive Monte Carlo simulations. Our results provide evidence that the previously reported nonzero $ε_\mathrm{c}$ is an artifact of a crossover phenomenon, which logarithmically converges to zero as the system size goes to infinity.

cond-mat.stat-mech↗