arXiv subjects

Wenyuan Yang

Publications and source records attributed to Wenyuan Yang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Conformal dynamics at infinity for groups with contracting elements

This paper develops a theory of conformal density at infinity for groups with contracting elements. We start by introducing a class of convergence boundary encompassing many known hyperbolic-like boundaries, on which a detailed study of conical points and Myrberg points is carried out. The basic theory of conformal density is then established on the convergence boundary, including the Sullivan shadow lemma and a Hopf--Tsuji--Sullivan dichotomy. This gives a unification of the theory of conformal density on the Gromov and Floyd boundary for (relatively) hyperbolic groups, the visual boundary for rank-1 CAT(0) groups, and Thurston boundary for mapping class groups. Besides that, the conformal density on the horofunction boundary provides a new important example of our general theory. Applications include the identification of Poisson boundary of random walks, the co-growth problem of divergent groups, measure theoretical results for CAT(0) groups and mapping class groups.

math.GR↗

arXiv2024

Uniform exponential growth for groups with proper product actions on hyperbolic spaces

This paper studies the locally uniform exponential growth and product set growth for a finitely generated group $G$ acting properly on a finite product of hyperbolic spaces. Under the assumption of coarsely dense orbits or shadowing property on factors, we prove that any finitely generated non-virtually abelian subgroup has uniform exponential growth. These assumptions are fulfilled in many hierarchically hyperbolic groups, including mapping class groups, specially cubulated groups and BMW groups. Moreover, if $G$ acts weakly acylindrically on each factor, we show that, with two exceptional classes of subgroups, $G$ has uniform product set growth. As corollaries, this gives a complete classification of subgroups with product set growth for any group acting discretely on a simply connected manifold with pinched negative curvature, for groups acting acylindrically on trees, and for 3-manifold groups.

math.GR↗

arXiv2024

Confined subgroups in groups with contracting elements

In this paper, we study the growth of confined subgroups through boundary actions of groups with contracting elements. We establish that the growth rate of a confined subgroup is strictly greater than half of the ambient growth rate in groups with purely exponential growth. Along the way, several results are obtained on the Hopf decomposition for boundary actions of subgroups with respect to conformal measures. In particular, we prove that confined subgroups are conservative, and examples of subgroups with nontrivial Hopf decomposition are constructed. We show a connection between Hopf decomposition and quotient growth and provide a dichotomy on quotient growth of Schreier graphs for subgroups in hyperbolic groups.

math.GR↗

arXiv2024

FedTracker: Furnishing Ownership Verification and Traceability for Federated Learning Model

Federated learning (FL) is a distributed machine learning paradigm allowing multiple clients to collaboratively train a global model without sharing their local data. However, FL entails exposing the model to various participants. This poses a risk of unauthorized model distribution or resale by the malicious client, compromising the intellectual property rights of the FL group. To deter such misbehavior, it is essential to establish a mechanism for verifying the ownership of the model and as well tracing its origin to the leaker among the FL participants. In this paper, we present FedTracker, the first FL model protection framework that provides both ownership verification and traceability. FedTracker adopts a bi-level protection scheme consisting of global watermark mechanism and local fingerprint mechanism. The former authenticates the ownership of the global model, while the latter identifies which client the model is derived from. FedTracker leverages Continual Learning (CL) principles to embed the watermark in a way that preserves the utility of the FL model on both primitive task and watermark task. FedTracker also devises a novel metric to better discriminate different fingerprints. Experimental results show FedTracker is effective in ownership verification, traceability, and maintains good fidelity and robustness against various watermark removal attacks.

cs.CR↗

arXiv2024

RobWE: Robust Watermark Embedding for Personalized Federated Learning Model Ownership Protection

Embedding watermarks into models has been widely used to protect model ownership in federated learning (FL). However, existing methods are inadequate for protecting the ownership of personalized models acquired by clients in personalized FL (PFL). This is due to the aggregation of the global model in PFL, resulting in conflicts over clients' private watermarks. Moreover, malicious clients may tamper with embedded watermarks to facilitate model leakage and evade accountability. This paper presents a robust watermark embedding scheme, named RobWE, to protect the ownership of personalized models in PFL. We first decouple the watermark embedding of personalized models into two parts: head layer embedding and representation layer embedding. The head layer belongs to clients' private part without participating in model aggregation, while the representation layer is the shared part for aggregation. For representation layer embedding, we employ a watermark slice embedding operation, which avoids watermark embedding conflicts. Furthermore, we design a malicious watermark detection scheme enabling the server to verify the correctness of watermarks before aggregating local models. We conduct an exhaustive experimental evaluation of RobWE. The results demonstrate that RobWE significantly outperforms the state-of-the-art watermark embedding schemes in FL in terms of fidelity, reliability, and robustness.

cs.CR↗

arXiv2023

Counting double cosets with application to generic 3-manifolds

We study the growth of double cosets in the class of groups with contracting elements, including relatively hyperbolic groups, CAT(0) groups and mapping class groups among others. Generalizing a recent work of Gitik and Rips about hyperbolic groups, we prove that the double coset growth of two Morse subgroups of infinite index is comparable with the orbital growth function. The same result is further obtained for a more general class of subgroups whose limit sets are proper subsets in the entire limit set of the ambient group. As an application, we confirm a conjecture of Maher that hyperbolic 3-manifolds are exponentially generic in the set of 3-manifolds built from Heegaard splitting using complexity in Teichmüller metric.

math.GR↗

arXiv2023

The growth of the Green function for random walks and Poincar{é} series

Given a probability measure $μ$ on a finitely generated group $Γ$, the Green function $G(x,y|r)$ encodes many properties of the random walk associated with $μ$. Finding asymptotics of $G(x,y|r)$ as $y$ goes to infinity is a common thread in probability theory and is usually referred as renewal theory in literature. Endowing $Γ$ with a word distance, we denote by $H_r(n)$ the sum of the Green function $G(e,x|r)$ along the sphere of radius $n$. This quantity appears naturally when studying asymptotic properties of branching random walks driven by $μ$ on $Γ$ and the behavior of $H_r(n)$ as $n$ goes to infinity is intimately related to renewal theory. Our motivation in this paper is to construct various examples of particular behaviors for $H_r(n)$. First, our main result exhibits a class of relatively hyperbolic groups with convergent Poincar{é} series generated by $H_r(n)$, which answers some questions raised in a previous paper of the authors. Along the way, we investigate the behavior of $H_r(n)$ for several classes of finitely generated groups, including abelian groups, certain nilpotent groups, lamplighter groups, and Cartesian products of free groups.

math.GR↗

arXiv2023

Inter-frame Accelerate Attack against Video Interpolation Models

Deep learning based video frame interpolation (VIF) method, aiming to synthesis the intermediate frames to enhance video quality, have been highly developed in the past few years. This paper investigates the adversarial robustness of VIF models. We apply adversarial attacks to VIF models and find that the VIF models are very vulnerable to adversarial examples. To improve attack efficiency, we suggest to make full use of the property of video frame interpolation task. The intuition is that the gap between adjacent frames would be small, leading to the corresponding adversarial perturbations being similar as well. Then we propose a novel attack method named Inter-frame Accelerate Attack (IAA) that initializes the perturbation as the perturbation for the previous adjacent frame and reduces the number of attack iterations. It is shown that our method can improve attack efficiency greatly while achieving comparable attack performance with traditional methods. Besides, we also extend our method to video recognition models which are higher level vision tasks and achieves great attack efficiency.

cs.CV↗

arXiv2023

FedZKP: Federated Model Ownership Verification with Zero-knowledge Proof

Federated learning (FL) allows multiple parties to cooperatively learn a federated model without sharing private data with each other. The need of protecting such federated models from being plagiarized or misused, therefore, motivates us to propose a provable secure model ownership verification scheme using zero-knowledge proof, named FedZKP. It is shown that the FedZKP scheme without disclosing credentials is guaranteed to defeat a variety of existing and potential attacks. Both theoretical analysis and empirical studies demonstrate the security of FedZKP in the sense that the probability for attackers to breach the proposed FedZKP is negligible. Moreover, extensive experimental results confirm the fidelity and robustness of our scheme.

cs.CR↗

arXiv2023

FedSOV: Federated Model Secure Ownership Verification with Unforgeable Signature

Federated learning allows multiple parties to collaborate in learning a global model without revealing private data. The high cost of training and the significant value of the global model necessitates the need for ownership verification of federated learning. However, the existing ownership verification schemes in federated learning suffer from several limitations, such as inadequate support for a large number of clients and vulnerability to ambiguity attacks. To address these limitations, we propose a cryptographic signature-based federated learning model ownership verification scheme named FedSOV. FedSOV allows numerous clients to embed their ownership credentials and verify ownership using unforgeable digital signatures. The scheme provides theoretical resistance to ambiguity attacks with the unforgeability of the signature. Experimental results on computer vision and natural language processing tasks demonstrate that FedSOV is an effective federated model ownership verification scheme enhanced with provable cryptographic security.

cs.CR↗

arXiv2023

MIS: A Multi-Identifier Management and Resolution System Based on Consortium Blockchain in Metaverse

The metaverse gradually evolves into a virtual world containing a series of interconnected sub-metaverses. Diverse digital resources, including identities, contents, services, and supporting data, are key components of the sub-metaverse. Therefore, a Domain Name System (DNS)-like system is necessary for efficient management and resolution. However, the legacy DNS was designed with security vulnerabilities and trust risks due to centralized issues. Blockchain is used to mitigate these concerns due to its decentralized features. Additionally, it supports identity management as a default feature, making it a natural fit for the metaverse. While there are several DNS alternatives based on the blockchain, they either manage only a single type of identifiers or isolate identities from other sorts of identifiers, making it difficult for sub-metaverses to coexist and connect with each other. This paper proposes a Multi-Identifier management and resolution System (MIS) in the metaverse, supporting the registration, resolution, and inter-translation functions. The basic MIS is portrayed as a four-tier architecture on a consortium blockchain due to its manageability, enhanced security, and efficiency properties. On-chain data is lightweight and compressed to save on storage while accelerating reading and writing operations. The resource data is encrypted based on the attributes of the sub-metaverse in the storage tier for privacy protection and access control. For users with decentralization priorities, a modification named EMIS is built on top of Ethereum. Finally, MIS is implemented on two testbeds and is available online as the open-source system. The first testbed consists of 4 physical servers located in the UK and Malaysia while the second is made up of 200 virtual machines (VMs) spread over 26 countries across all 5 continents on Google Cloud.

cs.NI↗

arXiv2023

Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring

Federated learning (FL) allows multiple participants to collaboratively build deep learning (DL) models without directly sharing data. Consequently, the issue of copyright protection in FL becomes important since unreliable participants may gain access to the jointly trained model. Application of homomorphic encryption (HE) in secure FL framework prevents the central server from accessing plaintext models. Thus, it is no longer feasible to embed the watermark at the central server using existing watermarking schemes. In this paper, we propose a novel client-side FL watermarking scheme to tackle the copyright protection issue in secure FL with HE. To our best knowledge, it is the first scheme to embed the watermark to models under the Secure FL environment. We design a black-box watermarking scheme based on client-side backdooring to embed a pre-designed trigger set into an FL model by a gradient-enhanced embedding method. Additionally, we propose a trigger set construction mechanism to ensure the watermark cannot be forged. Experimental results demonstrate that our proposed scheme delivers outstanding protection performance and robustness against various watermark removal attacks and ambiguity attack.

cs.CR↗

arXiv2023

Property (QT) for 3-manifold groups

According to Bestvina-Bromberg-Fujiwara, a finitely generated group is said to have property (QT) if it acts isometrically on a finite product of quasi-trees so that orbital maps are quasi-isometric embeddings. We prove that the fundamental group $π_1(M)$ of a compact, connected, orientable 3-manifold $M$ has property (QT) if and only if no summand in the sphere-disc decomposition of $M$ supports either Sol or Nil geometry. In particular, all compact, orientable, irreducible 3-manifold groups with nontrivial torus decomposition and not supporting Sol geometry have property (QT). In the course of our study, we establish property (QT) for the class of Croke-Kleiner admissible groups and of relatively hyperbolic groups under natural assumptions has property (QT).

math.GT↗

arXiv2022

Branching Random Walks on relatively hyperbolic groups

Let $Γ$ be a non-elementary relatively hyperbolic group with a finite generating set. Consider a finitely supported admissible and symmetric probability measure $μ$ on $Γ$ and a probability measure $ν$ on $\mathbb{N}$ with mean $r$. Let $\mathrm{BRW}(Γ,ν,μ)$ be the branching random walk on $Γ$ with offspring distribution $ν$ and base motion given by the random walk with step distribution $μ$. It is known that for $1 < r \leq R$ with $R$ the radius of convergence for the Green function of the random walk, the population of $\mathrm{BRW}(Γ,ν,μ)$ survives forever, but eventually vacates every finite subset of $Γ$. We prove that in this regime, the growth rate of the trace of the branching random walk is equal to the growth rate $ω_Γ(r)$ of the Green function of the underlying random walk. We also prove that the Hausdorff dimension of the limit set $Λ(r)$, which is the random subset of the Bowditch boundary consisting of all accumulation points of the trace of $\mathrm{BRW}(Γ,ν,μ)$, is equal to a constant times $ω_Γ(r)$.

math.PR↗

arXiv2022

A Secure and Efficient Data Deduplication Scheme with Dynamic Ownership Management in Cloud Computing

Encrypted data deduplication is an important technique for saving storage space and network bandwidth, which has been widely used in cloud storage. Recently, a number of schemes that solve the problem of data deduplication with dynamic ownership management have been proposed. However, these schemes suffer from low efficiency when the dynamic ownership changes a lot. To this end, in this paper, we propose a novel server-side deduplication scheme for encrypted data in a hybrid cloud architecture, where a public cloud (Pub-CSP) manages the storage and a private cloud (Pri-CSP) plays a role as the data owner to perform deduplication and dynamic ownership management. Further, to reduce the communication overhead we use an initial uploader check mechanism to ensure only the first uploader needs to perform encryption, and adopt an access control technique that verifies the validity of the data users before they download data. Our security analysis and performance evaluation demonstrate that our proposed server-side deduplication scheme has better performance in terms of security, effectiveness, and practicability compared with previous schemes. Meanwhile, our method can efficiently resist collusion attacks and duplicate faking attacks.

cs.CR↗

arXiv2021

The role of double ionization on the generation of doubly charged ions in copper vacuum arcs: insight from particle-in-cell/direct simulation Monte Carlo methods

Metal vapour vacuum arcs are capable to generate multiply charged metallic ions, which are widely used in fields such as ion deposition, ion thrusters, and ion sources, etc. According to the stationary model of cathode spot, those ions are generated by electron-impact single ionization in a step-wise manner, which is M -> M+ -> M2+ -> ... mainly. This paper is designed to study quantitatively the role of double ionization M -> M2+ in the breakdown initiation of copper vacuum arcs. A direct simulation Monte Carlo (DSMC) scheme of double ionization is proposed and incorporated into a 2D particle-in-cell (PIC) method. The super-particles of Cu2+ ions generated from different channels are labelled independently in the PIC-DSMC modelling of vacuum arc breakdown. The cathode erosion rate based on PIC modelling is about 40μg/C in arc burning regime, which agrees well with previous experiments. The temporal discharge behaviours such as arc current, arc voltage, and ionization degree of arc plasma, are influenced with or without double ionization negligibly. However, additional Cu2+ ions are generated near the cathode in breakdown initiation from the double ionization channel, with a lower kinetic energy on average. Therefore, the results on spatial distribution and energy spectra of Cu2+ ions are different with or without double ionization. This paper provides a quantitative research method to evaluate the role of multiply ionization in vacuum arcs.

physics.plasm-ph↗

arXiv2021

Croke-Kleiner admissible groups: Property (QT) and quasiconvexity

Croke-Kleiner admissible groups firstly introduced by Croke-Kleiner belong to a particular class of graph of groups which generalize fundamental groups of $3$--dimensional graph manifolds. In this paper, we show that if $G$ is a Croke-Kleiner admissible group, acting geometrically on a CAT(0) space $X$, then a finitely generated subgroup of $G$ has finite height if and only if it is strongly quasi-convex. We also show that if $G \curvearrowright X$ is a flip CKA action then $G$ is quasi-isometric embedded into a finite product of quasi-trees. With further assumption on the vertex groups of the flip CKA action $G \curvearrowright X$, we show that $G$ satisfies property (QT) that is introduced by Bestvina-Bromberg-Fujiwara.

math.GR↗

arXiv2020

The Hausdorff dimension of the harmonic measure for relatively hyperbolic groups

The paper studies the Hausdorff dimension of harmonic measures on various boundaries of a relatively hyperbolic group which are associated with random walks driven by a probability measure with finite first moment. With respect to the Floyd metric and the shortcut metric, we prove that the Hausdorff dimension of the harmonic measure equals the ratio of the entropy and the drift of the random walk. If the group is infinitely-ended, the same dimension formula is obtained for the end boundary endowed with a visual metric. In addition, the Hausdorff dimension of the visual metric is identified with the growth rate of the word metric. These results are complemented by a characterization of doubling visual metrics for accessible infinitely-ended groups : the visual metrics on the end boundary is doubling if and only if the group is virtually free. Consequently, there are at least two different bi-Hölder classes (and thus quasi-symmetric classes) of visual metrics on the end boundary.

math.GR↗