arXiv subjects

Ge Chen

Publications and source records attributed to Ge Chen.

At least 37 records · Page 2Linked to original sources

arXiv2025

A Provably Secure Network Protocol for Private Communication with Analysis and Tracing Resistance

Anonymous communication networks have emerged as crucial tools for obfuscating communication pathways and concealing user identities. However, their practical deployments face significant challenges, including susceptibility to artificial intelligence (AI)-powered metadata analysis, difficulties in decentralized architectures, and the absence of provable security guarantees. To address these issues, this paper proposes a novel decentralized anonymous routing protocol with resistance to tracing and traffic analysis. The protocol eliminates dependencies on the threshold model and trusted third-party setups, ensuring indistinguishable identity privacy even in highly adversarial environments. Different from traditional empirical security analysis of anonymous networks, this paper rigorously proves indistinguishable identity privacy for users even in extremely adversarial environments. Furthermore, simulations confirm its practical feasibility, demonstrating both security and efficiency. By achieving information sharing with privacy preservation, the proposed protocol offers a provably secure solution for privacy-preserving communication in digital environments.

cs.CR↗

arXiv2025

Superhorizon fluctuations and the cosmic dipole problem

Recent observations have identified a significant 4.9$σ$ tension between the cosmic dipole inferred from galaxy number counts and that derived from the Cosmic Microwave Background (CMB), suggesting a potential deviation from the cosmological principle. This work investigates whether superhorizon isocurvature perturbations in cold dark matter (CDM) can account for this discrepancy. We demonstrate that, unlike adiabatic modes which cancel at leading order, superhorizon isocurvature modes can generate an intrinsic CMB dipole without significantly affecting galaxy number counts, thereby explaining the observed mismatch. We explore both single-mode and continuous-spectrum cases, focusing on two concrete models: a nearly scale-invariant power-law spectrum with a UV cutoff and axion-induced isocurvature perturbations. For the axion scenario, we show that if the radial mode evolves during inflation, the resulting perturbations can match the required amplitude while evading current CMB constraints. Our analysis constrains the self-coupling of associated potential for the axion to the range $10^{-9} < λ< 4 \times 10^{-9}$. These findings offer a viable solution to the dipole tension and may serve as indirect evidence for axion dark matter.

astro-ph.CO↗

arXiv2025

Properties of Quasi-synchronization Time of High-dimensional Hegselmann-Krause Dynamics

The behavior of one-dimensional Hegselmann-Krause (HK) dynamics driven by noise has been extensively studied. Previous research has indicated that within no matter the bounded or the unbounded space of one dimension, the HK dynamics attain quasi-synchronization (synchronization in noisy case) in finite time. However, it remains unclear whether this phenomenon holds in high-dimensional space. This paper investigates the random time for quasi-synchronization of multi-dimensional HK model and reveals that the boundedness and dimensions of the space determine different outcomes. To be specific, if the space is bounded, quasi-synchronization can be attained almost surely for all dimensions within a finite time, whereas in unbounded space, quasi-synchronization can only be achieved in low-dimensional cases (one and two). Furthermore, different integrability of the random time of various cases is proved.

math.DS↗

arXiv2025

Towards Reward Fairness in RLHF: From a Resource Allocation Perspective

Rewards serve as proxies for human preferences and play a crucial role in Reinforcement Learning from Human Feedback (RLHF). However, if these rewards are inherently imperfect, exhibiting various biases, they can adversely affect the alignment of large language models (LLMs). In this paper, we collectively define the various biases present in rewards as the problem of reward unfairness. We propose a bias-agnostic method to address the issue of reward fairness from a resource allocation perspective, without specifically designing for each type of bias, yet effectively mitigating them. Specifically, we model preference learning as a resource allocation problem, treating rewards as resources to be allocated while considering the trade-off between utility and fairness in their distribution. We propose two methods, Fairness Regularization and Fairness Coefficient, to achieve fairness in rewards. We apply our methods in both verification and reinforcement learning scenarios to obtain a fairness reward model and a policy model, respectively. Experiments conducted in these scenarios demonstrate that our approach aligns LLMs with human preferences in a more fair manner.

cs.LG↗

arXiv2025

Neural Risk Limiting Dispatch in Power Networks: Formulation and Generalization Guarantees

Risk limiting dispatch (RLD) has been proposed as an approach that effectively trades off economic costs with operational risks for power dispatch under uncertainty. However, how to solve the RLD problem with provably near-optimal performance still remains an open problem. This paper presents a learning-based solution to this challenge. We first design a data-driven formulation for the RLD problem, which aims to construct a decision rule that directly maps day-ahead observable information to cost-effective dispatch decisions for the future delivery interval. Unlike most existing works that follow a predict-then-optimize paradigm, this end-to-end rule bypasses the additional suboptimality introduced by separately handling prediction and optimization. We then propose neural RLD, a novel solution method to the data-driven formulation. This method leverages an L2-regularized neural network to learn the decision rule, thereby transforming the data-driven formulation into a neural network training task that can be efficiently completed by stochastic gradient descent. A theoretical performance guarantee is further established to bound the suboptimality of our method, which implies that its suboptimality approaches zero with high probability as more samples are utilized. Simulation tests across various systems demonstrate our method's superior performance in convergence, suboptimality, and computational efficiency compared with benchmarks.

math.OC↗

arXiv2025

Noise-driven Synchronization of Vicsek Model in Mean

The Vicsek model has long stood as a pivotal framework in exploring collective behavior and self-organization, captivating the scientific community with its compelling dynamics. However, understanding how noise influences synchronization within this model and its associated phase transition characteristics has presented significant challenges. While numerous studies have focused on simulations due to the model's mathematical complexity, comprehensive theoretical analyses remain sparse. In this paper, we deliver a rigorous mathematical proof demonstrating that for any initial configuration of the Vicsek model, there exists a bound on noise amplitude such that if the noise amplitude is maintained within this bound, the system will achieve synchronization in mean. This finding not only lays a solid mathematical groundwork for the Vicsek model's phase transition theory but also underscores the critical role of noise in collective dynamics, enhancing our understanding of self-organizing systems in stochastic environments.

math-ph↗

arXiv2024

Multidimensional Opinion Dynamics with Heterogeneous Bounded Confidences and Random Interactions

This paper introduces a heterogeneous multidimensional bounded confidence (BC) opinion dynamics with random pairwise interactions, whereby each pair of agents accesses each other's opinions with a specific probability. This revised model is motivated by the observation that the standard Hegselmann-Krause (HK) dynamics requires unrealistic all-to-all interactions at certain configurations. For this randomized BC opinion dynamics, regardless of initial opinions and positive confidence bounds, we show that the agents' states converge to fixed final opinions in finite time almost surely and that the convergence rate follows a negative exponential distribution in mean square. Furthermore, we establish sufficient conditions for the heterogeneous BC opinion dynamics with random interactions to achieve consensus in finite time.

math.OC↗

arXiv2024

A "Breathing" Mobile Communication Network

The frequent migration of large-scale users leads to the load imbalance of mobile communication networks, which causes resource waste and decreases user experience. To address the load balancing problem, this paper proposes a dynamic optimization framework for mobile communication networks inspired by the average consensus in multi-agent systems. In this framework, all antennas cooperatively optimize their CPICH (Common Pilot Channel) transmit power in real-time to balance their busy-degrees. Then, the coverage area of each antenna would change accordingly, and we call this framework a ``breathing'' mobile communication network. To solve this optimization problem, two algorithms named BDBA (Busy-degree Dynamic Balancing Algorithm) and BFDBA (Busy-degree Fast Dynamic Balancing Algorithm) are proposed. Moreover, a fast network coverage calculation method is introduced, by which each antenna's minimum CPICH transmit power is determined under the premise of meeting the network coverage requirements. Besides, we present the theoretical analysis of the two proposed algorithms' performance, which prove that all antennas' busy-degrees will reach consensus under certain assumptions. Furthermore, simulations carried out on three large datasets demonstrate that our cooperative optimization can significantly reduce the unbalance among antennas as well as the proportion of over-busy antennas.

eess.SY↗

arXiv2024

GUNDAM: Aligning Large Language Models with Graph Understanding

Large Language Models (LLMs) have achieved impressive results in processing text data, which has sparked interest in applying these models beyond textual data, such as graphs. In the field of graph learning, there is a growing interest in harnessing LLMs to comprehend and manipulate graph-structured data. Existing research predominantly focuses on graphs with rich textual features, such as knowledge graphs or text attribute graphs, leveraging LLMs' ability to process text but inadequately addressing graph structure. This work specifically aims to assess and enhance LLMs' abilities to comprehend and utilize the structural knowledge inherent in graph data itself, rather than focusing solely on graphs rich in textual content. To achieve this, we introduce the \textbf{G}raph \textbf{U}nderstanding for \textbf{N}atural Language \textbf{D}riven \textbf{A}nalytical \textbf{M}odel (\model). This model adapts LLMs to better understand and engage with the structure of graph data, enabling them to perform complex reasoning tasks by leveraging the graph's structure itself. Our experimental evaluations on graph reasoning benchmarks not only substantiate that \model~ outperforms the SOTA baselines for comparisons. But also reveals key factors affecting the graph reasoning capabilities of LLMs. Moreover, we provide a theoretical analysis illustrating how reasoning paths can enhance LLMs' reasoning capabilities.

cs.AI↗

arXiv2024

Convergence of the Heterogeneous Deffuant-Weisbuch Model: A Complete Proof and Some Extensions

The Deffuant-Weisbuch (DW) model is a well-known bounded-confidence opinion dynamics that has attracted wide interest. Although the heterogeneous DW model has been studied by simulations over $20$ years, its convergence proof is open. Our previous paper \cite{GC-WS-WM-FB:20} solves the problem for the case of uniform weighting factors greater than or equal to $1/2$, but the general case remains unresolved. This paper considers the DW model with heterogeneous confidence bounds and heterogeneous (unconstrained) weighting factors and shows that, with probability one, the opinion of each agent converges to a fixed vector. In other words, this paper resolves the convergence conjecture for the heterogeneous DW model. Our analysis also clarifies how the convergence speed may be arbitrarily slow under certain parameter conditions.

math.OC↗

arXiv2024

Convergence Analysis of Weighted-Median Opinion Dynamics with Prejudice

The Friedkin-Johnsen (FJ) model introduces prejudice into the opinion evolution and has been successfully validated in many practical scenarios; however, due to its weighted average mechanism, only one prejudiced agent can always guide all unprejudiced agents synchronizing to its prejudice under the connected influence network, which may not be in line with some social realities. To fundamentally address the limitation of the weighted average mechanism, a weighted-median opinion dynamics has been recently proposed; however, its theoretical analysis is challenging due to its nonlinear nature. This paper studies the weighted-median opinion dynamics with prejudice, and obtains the convergence and convergence rate when all agents have prejudice, and a necessary and sufficient condition for asymptotic consensus when a portion of agents have prejudice. These results are the first time to analyze the discrete-time and synchronous opinion dynamics with the weighted median mechanism, and address the phenomenon of the FJ model that connectivity leads to consensus when a few agents with the same prejudice join in an unprejudiced group.

eess.SY↗

arXiv2024

An Optimal Pricing Formula for Smart Grid based on Stackelberg Game

The dynamic pricing of electricity is one of the most crucial demand response (DR) strategies in smart grid, where the utility company typically adjust electricity prices to influence user electricity demand. This paper models the relationship between the utility company and flexible electricity users as a Stackelberg game. Based on this model, we present a series of analytical results under certain conditions. First, we give an analytical Stackelberg equilibrium, namely the optimal pricing formula for utility company, as well as the unique and strict Nash equilibrium for users' electricity demand under this pricing scheme. To our best knowledge, it is the first optimal pricing formula in the research of price-based DR strategies. Also, if there exist prediction errors for the supply and demand of electricity, we provide an analytical expression for the energy supply cost of utility company. Moreover, a sufficient condition has been proposed that all electricity demands can be supplied by renewable energy. When the conditions for analytical results are not met, we provide a numerical solution algorithm for the Stackelberg equilibrium and verify its efficiency by simulation.

math.OC↗

arXiv2024

Preserving Node Distinctness in Graph Autoencoders via Similarity Distillation

Graph autoencoders (GAEs), as a kind of generative self-supervised learning approach, have shown great potential in recent years. GAEs typically rely on distance-based criteria, such as mean-square-error (MSE), to reconstruct the input graph. However, relying solely on a single reconstruction criterion may lead to a loss of distinctiveness in the reconstructed graph, causing nodes to collapse into similar representations and resulting in sub-optimal performance. To address this issue, we have developed a simple yet effective strategy to preserve the necessary distinctness in the reconstructed graph. Inspired by the knowledge distillation technique, we found that the dual encoder-decoder architecture of GAEs can be viewed as a teacher-student relationship. Therefore, we propose transferring the knowledge of distinctness from the raw graph to the reconstructed graph, achieved through a simple KL constraint. Specifically, we compute pairwise node similarity scores in the raw graph and reconstructed graph. During the training process, the KL constraint is optimized alongside the reconstruction criterion. We conducted extensive experiments across three types of graph tasks, demonstrating the effectiveness and generality of our strategy. This indicates that the proposed approach can be employed as a plug-and-play method to avoid vague reconstructions and enhance overall performance.

cs.LG↗

arXiv2024

Towards Comprehensive Preference Data Collection for Reward Modeling

Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models (LLMs) with human preferences, thereby enhancing the quality of responses generated. A critical component of RLHF is the reward model, which is trained on preference data and outputs a scalar reward during the inference stage. However, the collection of preference data still lacks thorough investigation. Recent studies indicate that preference data is collected either by AI or humans, where chosen and rejected instances are identified among pairwise responses. We question whether this process effectively filters out noise and ensures sufficient diversity in collected data. To address these concerns, for the first time, we propose a comprehensive framework for preference data collection, decomposing the process into four incremental steps: Prompt Generation, Response Generation, Response Filtering, and Human Labeling. This structured approach ensures the collection of high-quality preferences while reducing reliance on human labor. We conducted comprehensive experiments based on the data collected at different stages, demonstrating the effectiveness of the proposed data collection method.

cs.AI↗

arXiv2024

Production of doubly charmed hadron $Ξ_{cc}^{++}$ and $T_{cc}^+$ in relativistic heavy ion collisions

Heavy ion collisions provide a unique opportunity for studying the properties of exotic hadrons with two charm quarks. The production of $T_{cc}^+$ is significantly enhanced in nuclear collisions compared to proton-proton collisions due to the creation of multiple charm pairs. In this study, we employ the Langevin equation in combination with the Instantaneous Coalescence Model (LICM) to investigate the production of $T_{cc}^+$ and $Ξ_{cc}^{++}$ which consists of two charm quarks. We consider $T_{cc}^+$ as molecular states composed of $D$ and $D^*$ mesons. The Langevin equation is used to calculate the energy loss of charm quarks and $D$ mesons in the hot medium. The hadronization process, where charm quarks transform into each $D$ state as constituents of $T_{cc}^+$ production, is described using the coalescence model. The coalescence probability between $D$ and $D^*$ is determined by the Wigner function, which encodes the information of the $T_{cc}^+$ wave function. Our results show that the $T_{cc}^+$ production varies by approximately one order of magnitude when different widths in the Wigner function, representing distinct binding energies of $T_{cc}^+$, are considered. This variation offers valuable insights into the nature of $T_{cc}^+$ through the analysis of its wave function. The $Ξ_{cc}^{++}$ is treated as a hadronic state produced at the hadronization of the deconfined matter. Its production is also calculated as a comparison with the molecular state $T_{cc}^+$.

nucl-th↗

arXiv2024

Graph Ranking Contrastive Learning: A Extremely Simple yet Efficient Method

Graph contrastive learning (GCL) has emerged as a representative graph self-supervised method, achieving significant success. The currently prevalent optimization objective for GCL is InfoNCE. Typically, it employs augmentation techniques to obtain two views, where a node in one view acts as the anchor, the corresponding node in the other view serves as the positive sample, and all other nodes are regarded as negative samples. The goal is to minimize the distance between the anchor node and positive samples and maximize the distance to negative samples. However, due to the lack of label information during training, InfoNCE inevitably treats samples from the same class as negative samples, leading to the issue of false negative samples. This can impair the learned node representations and subsequently hinder performance in downstream tasks. While numerous methods have been proposed to mitigate the impact of false negatives, they still face various challenges. For instance, while increasing the number of negative samples can dilute the impact of false negatives, it concurrently increases computational burden. Thus, we propose GraphRank, a simple yet efficient graph contrastive learning method that addresses the problem of false negative samples by redefining the concept of negative samples to a certain extent, thereby avoiding the issue of false negative samples. The effectiveness of GraphRank is empirically validated through experiments on the node, edge, and graph level tasks.

cs.LG↗

arXiv2024

Exploring Task Unification in Graph Representation Learning via Generative Approach

Graphs are ubiquitous in real-world scenarios and encompass a diverse range of tasks, from node-, edge-, and graph-level tasks to transfer learning. However, designing specific tasks for each type of graph data is often costly and lacks generalizability. Recent endeavors under the "Pre-training + Fine-tuning" or "Pre-training + Prompt" paradigms aim to design a unified framework capable of generalizing across multiple graph tasks. Among these, graph autoencoders (GAEs), generative self-supervised models, have demonstrated their potential in effectively addressing various graph tasks. Nevertheless, these methods typically employ multi-stage training and require adaptive designs, which on one hand make it difficult to be seamlessly applied to diverse graph tasks and on the other hand overlook the negative impact caused by discrepancies in task objectives between the different stages. To address these challenges, we propose GA^2E, a unified adversarially masked autoencoder capable of addressing the above challenges seamlessly. Specifically, GA^2E proposes to use the subgraph as the meta-structure, which remains consistent across all graph tasks (ranging from node-, edge-, and graph-level to transfer learning) and all stages (both during training and inference). Further, GA^2E operates in a \textbf{"Generate then Discriminate"} manner. It leverages the masked GAE to reconstruct the input subgraph whilst treating it as a generator to compel the reconstructed graphs resemble the input subgraph. Furthermore, GA^2E introduces an auxiliary discriminator to discern the authenticity between the reconstructed (generated) subgraph and the input subgraph, thus ensuring the robustness of the graph representation through adversarial training mechanisms. We validate GA^2E's capabilities through extensive experiments on 21 datasets across four types of graph tasks.

cs.LG↗

arXiv2024

Bottomonium evolution with in-medium heavy quark potential from lattice QCD

The static properties and dynamic evolution of bottomonium states in a hot QCD medium are investigated through the Schrödinger equation with complex heavy quark potentials, which are presented recently in lattice QCD study and with three different extractions. This approach builds a direct connection between the in-medium heavy quark potentials from the lattice QCD to the experimental observables. The yields and nuclear modification factors $R_{AA}$ of bottomonium in Pb-Pb collisions at $\sqrt{s_{\rm NN}}=5.02~\rm TeV$ are calculated in this work. Our results show a large suppression of the bottomonium yield in heavy ion collisions due to the large imaginary potential. To understand bottomonium $R_{AA}$ based on lattice QCD potentials, we propose a fomration time for bottomonium states and find that experimental data can be well explained with the heavy quark potential extracted by the Padé fit, which shows no color screening in the real part potential.

nucl-th↗