arXiv subjects

Hawoong Jeong

Publications and source records attributed to Hawoong Jeong.

At least 19 recordsLinked to original sources

arXiv2026

Exact Identity Linking Entropy Production and Mutual Information

We establish an exact identity for overdamped Langevin dynamics: the total entropy production rate equals four times the mutual information rate between an infinitesimal displacement and its actual temporal midpoint, plus a mean-flow term. This provides a forward information-theoretic representation of irreversibility and makes the Shannon chain rule act directly on dissipation. For additive bipartitions, the chain rule reveals nonnegative self and interaction channels, with the self channel coinciding with apparent entropy production. This channel structure sharpens the learning-rate bound by showing that learning is constrained solely by the interaction channel. The same structure is also shown to resolve a recently reported coarse-graining reversal in hydrodynamically coupled colloids. Together, these results exhibit that, within this channel decomposition, learning and coarse-graining probe complementary parts of dissipation in overdamped Langevin systems.

cond-mat.stat-mech↗

arXiv2026

Role of volatility mixing in wealth condensation transition

We study the role of heterogeneous volatility in a networked wealth dynamics model and its impact on the wealth condensation transition. Extending the Bouchaud-Mézard framework, we introduce binary volatility in networks and investigate how its configuration affects the effective power-law tail exponent of the wealth distribution. Using a stochastic block model, we control the mixing between volatility groups and show that the effective exponent is governed not only by the global parameter $Λ=2J/β^2$ but also by the volatility configuration in the network. We find that local interactions between nodes with different volatility induce the neutralization of group-wise exponents, which lowers the aggregate tail exponent and yields a condensation transition across $γ_{\rm c}=2$. Our results identify volatility mixing as another control mechanism for wealth condensation and highlight the importance of noise heterogeneity in nonequilibrium systems on networks.

cond-mat.stat-mech↗

arXiv2026

Dataset Complexity Shapes Finite-Distance Loss Geometry in Neural Networks

Finite datasets can share the same size and low-order statistics while differing strongly in structural complexity. We connect this dataset complexity to loss-landscape geometry by pairing local label mixing across neighborhood scales with local entropy around trained neural-network solutions. Adapted from the Franz--Parisi construction in spin-glass theory, local entropy measures the effective volume of low-loss, solution-like parameter configurations at each distance from a reference. We estimate it in finite networks using adaptive sequential Monte Carlo. In a controlled synthetic sweep, greater dataset complexity produces a larger decrease in local entropy near the reference. Farther away, its radial derivative becomes weak and nearly common across conditions. Dataset complexity therefore changes where the effective solution volume contracts, rather than making it decrease uniformly faster. Experiments on real image data show the same qualitative trend, with label randomization further amplifying the effect. These results show that dataset structure shapes how low-loss neighborhoods are organized across finite distances from trained solutions.

cond-mat.dis-nn↗

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

Uncovering Spontaneous Physics Representations in In-Context Learning

In-context learning (ICL) lets large language models (LLMs) solve new tasks from prompts alone, across an ever-widening range of domains, yet the mechanisms underlying this ability remain poorly understood. Physical systems offer a controlled testbed for this question as they provide experimentally controllable data with structured dynamics grounded in fundamental principles. Here we study the ICL ability of LLMs, focusing on physical reasoning. Using dynamics forecasting as a proxy task, we first show that LLMs forecast physical dynamics in context, with accuracy improving as more history is provided. Analyzing the model's residual stream reveals internal activations that correlate with key physical quantities such as energy. These correlations strengthen gradually with context length, indicating that LLMs spontaneously form representations aligned with physical concepts without any physics-specific supervision. To assess whether these representations contribute to the model's predictions, we introduce a layer-wise gradient-based attribution analysis. We find that, residual directions more strongly correlated with energy also receive greater attribution to numerical predictions. This pattern is not observed for features correlated with directly observed quantities such as displacement, suggesting that the energy-related signal is not merely numerical information copied from the input. Our results broaden ICL analysis to structured physical dynamics and give a mechanistic account of how LLMs organize physical structure in context.

cs.CL↗

arXiv2026

Opinion Formation in a Spatially Constrained Coevolving Nonlinear Voter Model

We investigate a spatially constrained coevolving nonlinear voter model. Using a random geometric graph, we constrain the interaction range of voter dynamics. If local rewiring is not possible, the discordant link is deleted. Our results reveal absorbing states that differ not only in magnetization and activity, but also in mean degree and spatial state organization. By exploring dynamical and structural observables, we found that distinct regimes from the existing coevolving nonlinear voter model are characterized by a reduced consensus region, spatially segregated fragmentation, and isolated node formation. Additionally, we develop a phenomenological description of the evolution of the mean degree and terminal non-conserved quantities that is consistent with the numerical results. Our model highlights how local geometric accessibility reshapes the structure of the absorbing states.

cond-mat.stat-mech↗

arXiv2026

Impact of capacity volatility and input substitutability on supply chain resilience

Supply chains are intrinsically vulnerable to stochastic shocks due to their sequential production dependencies. Building on the Feld-Barthelemy framework, we investigate how capacity volatility and input substitutability determine critical demands in stochastic supply chains. By modeling production capacity with a truncated normal distribution, we show that in long supply chains, reducing capacity volatility is often more effective than increasing average capacity, emphasizing the need for firm-level synchronization. Furthermore, introducing a modified Leontief-type production function reveals that input substitutability effectively disperses stochastic shocks. Supplier diversification inherently raises critical demands, even under fixed maximum capacities, by introducing the effect of network topology that independently enhances the resilience of physical stock. Our findings demonstrate that mitigating capacity volatility and structurally diversifying supply routes are just as crucial to supply chain resilience as traditional inventory expansion.

cond-mat.stat-mech↗

arXiv2026

Anomaly, class division, and decoupling in income dynamics

Economic inequality emerges from the interplay between regional growth-rate differences and the interaction network that couples regions. We propose a minimal income-dynamics model, where heterogeneity is governed by growth-rate assortativity $\mathcal{A}$ and regional concentration $\mathcal{R}$, allowing us to quantify the spatiotemporal patterns of empirically observed log-income distributions. To systematically analyze these patterns, we derive closed-form approximations for the Hellinger distance and the Gini index in limiting configurations. Our findings highlight the spatial segregation of growth rates as a key driver of economic class division and demonstrate how small-world shortcuts in the underlying network can disrupt this segregation. Finally, our framework provides a robust explanation for the bimodality and strong regional correlations found in global income distributions.

cond-mat.stat-mech↗

arXiv2025

Gaussian Universality in Neural Network Dynamics with Generalized Structured Input Distributions

Analyzing neural network dynamics via stochastic gradient descent (SGD) is crucial to building theoretical foundations for deep learning. Previous work has analyzed structured inputs within the \textit{hidden manifold model}, often under the simplifying assumption of a Gaussian distribution. We extend this framework by modeling inputs as Gaussian mixtures to better represent complex, real-world data. Through empirical and theoretical investigation, we demonstrate that with proper standardization, the learning dynamics converges to the behavior seen in the simple Gaussian case. This finding exhibits a form of universality, where diverse structured distributions yield results consistent with Gaussian assumptions, thereby strengthening the theoretical understanding of deep learning models.

stat.ML↗

arXiv2025

Network analysis reveals news press landscape and asymmetric user polarization

Unlike traditional media, online news platforms allow users to consume content that suits their tastes and to facilitate interactions with other people. However, as more personalized consumption of information and interaction with like-minded users increase, ideological bias can inadvertently increase and contribute to the formation of echo chambers, reinforcing the polarization of opinions. Although the structural characteristics of polarization among different ideological groups in online spaces have been extensively studied, research into how these groups emotionally interact with each other has not been as thoroughly explored. From this perspective, we investigate both structural and affective polarization between news media user groups on Naver News, South Korea's largest online news portal, during the period of 2022 Korean presidential election. By utilizing the dataset comprising 333,014 articles and over 36 million user comments, we uncover two distinct groups of users characterized by opposing political leanings and reveal significant bias and polarization among them. Additionally, we reveal the existence of echo chambers within co-commenting networks and investigate the asymmetric affective interaction patterns between the two polarized groups. Classification task of news media articles based on the distinct comment response patterns support the notion that different political groups may employ distinct communication strategies. Our approach based on network analysis on large-scale comment dataset offers novel insights into characteristics of user polarization in the online news platforms and the nuanced interaction nature between user groups.

cs.SI↗

arXiv2025

Quantitative evaluation of methods to analyze motion changes in single-particle experiments

The analysis of live-cell single-molecule imaging experiments can reveal valuable information about the heterogeneity of transport processes and interactions between cell components. These characteristics are seen as motion changes in the particle trajectories. Despite the existence of multiple approaches to carry out this type of analysis, no objective assessment of these methods has been performed so far. Here, we report the results of a competition to characterize and rank the performance of these methods when analyzing the dynamic behavior of single molecules. To run this competition, we implemented a software library that simulates realistic data corresponding to widespread diffusion and interaction models, both in the form of trajectories and videos obtained in typical experimental conditions. The competition constitutes the first assessment of these methods, providing insights into the current limitations of the field, fostering the development of new approaches, and guiding researchers to identify optimal tools for analyzing their experiments.

cond-mat.soft↗

arXiv2025

Inferring the Langevin Equation with Uncertainty via Bayesian Neural Networks

Pervasive across diverse domains, stochastic systems exhibit fluctuations in processes ranging from molecular dynamics to climate phenomena. The Langevin equation has served as a common mathematical model for studying such systems, enabling predictions of their temporal evolution and analyses of thermodynamic quantities, including absorbed heat, work done on the system, and entropy production. However, inferring the Langevin equation from observed trajectories is a challenging problem, and assessing the uncertainty associated with the inferred equation has yet to be accomplished. In this study, we present a comprehensive framework that employs Bayesian neural networks for inferring Langevin equations in both overdamped and underdamped regimes. Our framework first provides the drift force and diffusion matrix separately and then combines them to construct the Langevin equation. By providing a distribution of predictions instead of a single value, our approach allows us to assess prediction uncertainties, which can help prevent potential misunderstandings and erroneous decisions about the system. We demonstrate the effectiveness of our framework in inferring Langevin equations for various scenarios including a neuron model and microscopic engine, highlighting its versatility and potential impact.

cond-mat.stat-mech↗

arXiv2025

Stochastic Resetting Mitigates Latent Gradient Bias of SGD from Label Noise

Giving up and starting over may seem wasteful in many situations such as searching for a target or training deep neural networks (DNNs). Our study, though, demonstrates that resetting from a checkpoint can significantly improve generalization performance when training DNNs with noisy labels. In the presence of noisy labels, DNNs initially learn the general patterns of the data but then gradually memorize the corrupted data, leading to overfitting. By deconstructing the dynamics of stochastic gradient descent (SGD), we identify the behavior of a latent gradient bias induced by noisy labels, which harms generalization. To mitigate this negative effect, we apply the stochastic resetting method to SGD, inspired by recent developments in the field of statistical physics achieving efficient target searches. We first theoretically identify the conditions where resetting becomes beneficial, and then we empirically validate our theory, confirming the significant improvements achieved by resetting. We further demonstrate that our method is both easy to implement and compatible with other methods for handling noisy labels. Additionally, this work offers insights into the learning dynamics of DNNs from an interpretability perspective, expanding the potential to analyze training methods through the lens of statistical physics.

cs.LG↗

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

Interplay of network structure and talent configuration on wealth dynamics

The economic success of individuals is often determined by a combination of talent, luck, and assistance from others. We introduce a new agent-based model that simultaneously considers talent, luck, and social interaction. This model allows us to explore how network structure (how agents interact) and talent distribution among agents affect the dynamics of capital accumulation through analytical and numerical methods. We identify a phenomenon as ``talent configuration effect", which refers to the influence of how talent is allocated to individuals (nodes) in the network. We analyze this effect through two key properties: talent assortativity (TA) and talent-degree correlation (TD). In particular, we focus on three economic indicators: growth rate ($n_{\rm rate}$), Gini coefficient (inequality: $n_{\rm Gini}$), and meritocratic fairness ($n_{LT}$). This investigation helps us understand the interplay between talent configuration and network structure on capital dynamics. We find that, in the short term, positive correlations exist between TA and TD for all three economic indicators. Furthermore, the dominant factor influencing capital dynamics depends on the network topology. In scale-free networks, TD has a stronger influence on the economic indices than TA. Conversely, in lattice-like networks, TA plays a more significant role. Our findings address that high socioeconomic homophily can create a dilemma between growth and equality, and that hub monopolization by few highly talented agents makes economic growth strongly dependent on their performances.

physics.soc-ph↗

arXiv2024

Towards Cross Domain Generalization of Hamiltonian Representation via Meta Learning

Recent advances in deep learning for physics have focused on discovering shared representations of target systems by incorporating physics priors or inductive biases into neural networks. While effective, these methods are limited to the system domain, where the type of system remains consistent and thus cannot ensure the adaptation to new, or unseen physical systems governed by different laws. For instance, a neural network trained on a mass-spring system cannot guarantee accurate predictions for the behavior of a two-body system or any other system with different physical laws. In this work, we take a significant leap forward by targeting cross domain generalization within the field of Hamiltonian dynamics. We model our system with a graph neural network (GNN) and employ a meta learning algorithm to enable the model to gain experience over a distribution of systems and make it adapt to new physics. Our approach aims to learn a unified Hamiltonian representation that is generalizable across multiple system domains, thereby overcoming the limitations of system-specific models. We demonstrate that the meta-trained model captures the generalized Hamiltonian representation that is consistent across different physical domains. Overall, through the use of meta learning, we offer a framework that achieves cross domain generalization, providing a step towards a unified model for understanding a wide array of dynamical systems via deep learning.

cs.LG↗

arXiv2024

Hidden multiscale organization and robustness of real multiplex networks

Hidden geometry enables the investigation of complex networks at different scales. Extending this framework to multiplex networks, we uncover a novel kind of mesoscopic organization in real multiplex systems, named $\textit{clan}$, a group of nodes that preserve their local geometric arrangement across layers. Furthermore, we reveal the intimate relationship between the unfolding of clan structure and mutual percolation against targeted attacks, leading to an ambivalent role of clans: making a system fragile yet less prone to complete shattering. Finally, we confirm the correlation between the multiscale nature of geometric organization and the overall robustness. Our findings expand the significance of hidden geometry in network function, while also highlighting potential pitfalls in evaluating and controlling catastrophic failure of multiplex systems.

physics.soc-ph↗

arXiv2023

Social learning spontaneously emerges by searching optimal heuristics with deep reinforcement learning

How have individuals of social animals in nature evolved to learn from each other, and what would be the optimal strategy for such learning in a specific environment? Here, we address both problems by employing a deep reinforcement learning model to optimize the social learning strategies (SLSs) of agents in a cooperative game in a multi-dimensional landscape. Throughout the training for maximizing the overall payoff, we find that the agent spontaneously learns various concepts of social learning, such as copying, focusing on frequent and well-performing neighbors, self-comparison, and the importance of balancing between individual and social learning, without any explicit guidance or prior knowledge about the system. The SLS from a fully trained agent outperforms all of the traditional, baseline SLSs in terms of mean payoff. We demonstrate the superior performance of the reinforcement learning agent in various environments, including temporally changing environments and real social networks, which also verifies the adaptability of our framework to different social settings.

cs.LG↗