arXiv subjects

Yiwei Zhang

Publications and source records attributed to Yiwei Zhang.

At least 55 records · Page 3Linked to original sources

arXiv2025

Dual-Modality Computational Ophthalmic Imaging with Deep Learning and Coaxial Optical Design

The growing burden of myopia and retinal diseases necessitates more accessible and efficient eye screening solutions. This study presents a compact, dual-function optical device that integrates fundus photography and refractive error detection into a unified platform. The system features a coaxial optical design using dichroic mirrors to separate wavelength-dependent imaging paths, enabling simultaneous alignment of fundus and refraction modules. A Dense-U-Net-based algorithm with customized loss functions is employed for accurate pupil segmentation, facilitating automated alignment and focusing. Experimental evaluations demonstrate the system's capability to achieve high-precision pupil localization (EDE = 2.8 px, mIoU = 0.931) and reliable refractive estimation with a mean absolute error below 5%. Despite limitations due to commercial lens components, the proposed framework offers a promising solution for rapid, intelligent, and scalable ophthalmic screening, particularly suitable for community health settings.

eess.IV↗

arXiv2025

Addressing Class Imbalance with Probabilistic Graphical Models and Variational Inference

This study proposes a method for imbalanced data classification based on deep probabilistic graphical models (DPGMs) to solve the problem that traditional methods have insufficient learning ability for minority class samples. To address the classification bias caused by class imbalance, we introduce variational inference optimization probability modeling, which enables the model to adaptively adjust the representation ability of minority classes and combines the class-aware weight adjustment strategy to enhance the classifier's sensitivity to minority classes. In addition, we combine the adversarial learning mechanism to generate minority class samples in the latent space so that the model can better characterize the category boundary in the high-dimensional feature space. The experiment is evaluated on the Kaggle "Credit Card Fraud Detection" dataset and compared with a variety of advanced imbalanced classification methods (such as GAN-based sampling, BRF, XGBoost-Cost Sensitive, SAAD, HAN). The results show that the method in this study has achieved the best performance in AUC, Precision, Recall and F1-score indicators, effectively improving the recognition rate of minority classes and reducing the false alarm rate. This method can be widely used in imbalanced classification tasks such as financial fraud detection, medical diagnosis, and anomaly detection, providing a new solution for related research.

cs.LG↗

arXiv2025

Federated Learning for Cross-Domain Data Privacy: A Distributed Approach to Secure Collaboration

This paper proposes a data privacy protection framework based on federated learning, which aims to realize effective cross-domain data collaboration under the premise of ensuring data privacy through distributed learning. Federated learning greatly reduces the risk of privacy breaches by training the model locally on each client and sharing only model parameters rather than raw data. The experiment verifies the high efficiency and privacy protection ability of federated learning under different data sources through the simulation of medical, financial, and user data. The results show that federated learning can not only maintain high model performance in a multi-domain data environment but also ensure effective protection of data privacy. The research in this paper provides a new technical path for cross-domain data collaboration and promotes the application of large-scale data analysis and machine learning while protecting privacy.

cs.LG↗

arXiv2025

A Deep Learning Approach to Anomaly Detection in High-Frequency Trading Data

This paper proposes an algorithm based on a staged sliding window Transformer architecture to detect abnormal behaviors in the microstructure of the foreign exchange market, focusing on high-frequency EUR/USD trading data. The method captures multi-scale temporal features through a staged sliding window, extracts global and local dependencies by combining the self-attention mechanism and weighted attention mechanism of the Transformer, and uses a classifier to identify abnormal events. Experimental results on a real high-frequency dataset containing order book depth, spread, and trading volume show that the proposed method significantly outperforms traditional machine learning (such as decision trees and random forests) and deep learning methods (such as MLP, CNN, RNN, LSTM) in terms of accuracy (0.93), F1-Score (0.91), and AUC-ROC (0.95). Ablation experiments verify the contribution of each component, and the visualization of order book depth and anomaly detection further reveals the effectiveness of the model under complex market dynamics. Despite the false positive problem, the model still provides important support for market supervision. In the future, noise processing can be optimized and extended to other markets to improve generalization and real-time performance.

cs.LG↗

arXiv2025

Social Network User Profiling for Anomaly Detection Based on Graph Neural Networks

This study proposes a risk pricing anomaly detection method for social network user portraits based on graph neural networks (GNNs), aiming to improve the ability to identify abnormal users in social network environments. In view of the limitations of traditional methods in social network data modeling, this paper combines graph autoencoders (GAEs) and graph attention networks (GATs) to achieve accurate detection of abnormal users through dynamic aggregation of neighbor features and reconstruction error evaluation. The Facebook Page-Page Network dataset is used in the experiment and compared with VAE, GNN, Transformer and GAE. The results show that the proposed method achieves the best performance in AUC, F1-score, Precision and Recall, verifying its effectiveness. In addition, this paper explores the computational efficiency of the model in large-scale data and looks forward to combining self-supervised learning, federated learning, and other technologies in the future to improve the robustness and privacy protection of risk assessment. The research results can provide efficient anomaly detection solutions for financial risk control, social security management, and other fields.

cs.LG↗

arXiv2025

Context-Aware Rule Mining Using a Dynamic Transformer-Based Framework

This study proposes a dynamic rule data mining algorithm based on an improved Transformer architecture, aiming to improve the accuracy and efficiency of rule mining in a dynamic data environment. With the increase in data volume and complexity, traditional data mining methods are difficult to cope with dynamic data with strong temporal and variable characteristics, so new algorithms are needed to capture the temporal regularity in the data. By improving the Transformer architecture, and introducing a dynamic weight adjustment mechanism and a temporal dependency module, we enable the model to adapt to data changes and mine more accurate rules. Experimental results show that compared with traditional rule mining algorithms, the improved Transformer model has achieved significant improvements in rule mining accuracy, coverage, and stability. The contribution of each module in the algorithm performance is further verified by ablation experiments, proving the importance of temporal dependency and dynamic weight adjustment mechanisms in improving the model effect. In addition, although the improved model has certain challenges in computational efficiency, its advantages in accuracy and coverage enable it to perform well in processing complex dynamic data. Future research will focus on optimizing computational efficiency and combining more deep learning technologies to expand the application scope of the algorithm, especially in practical applications in the fields of finance, medical care, and intelligent recommendation.

cs.LG↗

arXiv2025

Constrained Coding for Composite DNA: Channel Capacity and Efficient Constructions

Composite DNA is a recent novel method to increase the information capacity of DNA-based data storage above the theoretical limit of 2 bits/symbol. In this method, every composite symbol does not store a single DNA nucleotide but a mixture of the four nucleotides in a predetermined ratio. By using different mixtures and ratios, the alphabet can be extended to have much more than four symbols in the naive approach. While this method enables higher data content per synthesis cycle, potentially reducing the DNA synthesis cost, it also imposes significant challenges for accurate DNA sequencing since the base-level errors can easily change the mixture of bases and their ratio, resulting in changes to the composite symbols. With this motivation, we propose efficient constrained coding techniques to enforce the biological constraints, including the runlength-limited constraint and the GC-content constraint, into every DNA synthesized oligo, regardless of the mixture of bases in each composite letter and their corresponding ratio. Our goals include computing the capacity of the constrained channel, constructing efficient encoders/decoders, and providing the best options for the composite letters to obtain capacity-approaching codes. For certain codes' parameters, our methods incur only one redundant symbol.

cs.IT↗

arXiv2025

Asymptotically Optimal Codes for $(t,s)$-Burst Error

Recently, codes for correcting a burst of errors have attracted significant attention. One of the most important reasons is that bursts of errors occur in certain emerging techniques, such as DNA storage. In this paper, we investigate a type of error, called a $(t,s)$-burst, which deletes $t$ consecutive symbols and inserts $s$ arbitrary symbols at the same coordinate. Note that a $(t,s)$-burst error can be seen as a generalization of a burst of insertions ($t=0$), a burst of deletions ($s=0$), and a burst of substitutions ($t=s$). Our main contribution is to give explicit constructions of $q$-ary $(t,s)$-burst correcting codes with $\log n + O(1)$ bits of redundancy for any given constant non-negative integers $t$, $s$, and $q \geq 2$. These codes have optimal redundancy up to an additive constant. Furthermore, we apply our $(t,s)$-burst correcting codes to combat other various types of errors and improve the corresponding results. In particular, one of our byproducts is a permutation code capable of correcting a burst of $t$ stable deletions with $\log n + O(1)$ bits of redundancy, which is optimal up to an additive constant.

cs.IT↗

arXiv2024

Ground states and periodic orbits for expanding Thurston maps

Expanding Thurston maps form a class of branched covering maps on the topological $2$-sphere $S^{2}$, which are topological models of some non-uniformly expanding rational maps without any smoothness or holomorphicity assumption initially investigated by W. P. Thurston, M. Bonk, D. Meyer, P. Haïssinsky, and K. M. Pilgrim. The measures of maximal entropy and the absolutely continuous invariant measures for these maps have been studied by these authors, and equilibrium states by the first-named author. In this paper, we initiate the investigation on two new classes of invariant measures, namely, the maximizing measures and ground states, and establish the Livšic theorem, a local Anosov closing lemma, and give a positive answer to the Typically Periodic Optimization Conjecture from ergodic optimization for these maps. As an application, we establish these results for Misiurewicz--Thurston rational maps (i.e., postcritically-finite rational maps without periodic critical points) on the Riemann sphere including the Lattès maps with respect to the spherical metric. Our strategy relies on the visual metrics developed by the above authors. In particular, we verify, in a first non-uniformly expanding setting, the Typically Periodic Optimization Conjecture, establishing that for a generic Hölder continuous potential, there exists a unique maximizing measure, moreover, this measure is supported on a periodic orbit, it satisfies the locking property, and it is the unique ground state. The expanding Thurston maps we consider include those that are not topologically conjugate to rational maps; in particular, they can have periodic critical points.

math.DS↗

arXiv2024

Thermodynamic formalism for subsystems of expanding Thurston maps and large deviations asymptotics

Expanding Thurston maps were introduced by M. Bonk and D. Meyer with motivation from complex dynamics and Cannon's conjecture from geometric group theory via Sullivan's dictionary. In this paper, we introduce subsystems of expanding Thurston maps motivated via Sullivan's dictionary as analogs of some subgroups of Kleinian groups. We use thermodynamic formalism to prove the Variational Principle and the existence of equilibrium states for strongly irreducible subsystems and real-valued Hölder continuous potentials. Here, the sphere $S^{2}$ is equipped with a natural metric, called a visual metric, introduced by M. Bonk and D. Meyer. As an application, we establish large deviation asymptotics for expanding Thurston maps.

math.DS↗

arXiv2024

Bifurcations and invariant sets for a family of replicator maps from evolutionary games

We study the dynamics of a family of replicator maps, depending on two parameters. Such studies are motivated by the analysis of the dynamics of evolutionary games under selections. From the dynamics viewpoint, we prove the existence of hyperbolic chaos for the considered map. Moreover, we also give a partial solution of an open problem formulated in \cite{Misiurewicz1}: to describe all the one-dimensional maps with all the periodic orbits having the same mean value.

math.DS↗

arXiv2024

VQ-Map: Bird's-Eye-View Map Layout Estimation in Tokenized Discrete Space via Vector Quantization

Bird's-eye-view (BEV) map layout estimation requires an accurate and full understanding of the semantics for the environmental elements around the ego car to make the results coherent and realistic. Due to the challenges posed by occlusion, unfavourable imaging conditions and low resolution, \emph{generating} the BEV semantic maps corresponding to corrupted or invalid areas in the perspective view (PV) is appealing very recently. \emph{The question is how to align the PV features with the generative models to facilitate the map estimation}. In this paper, we propose to utilize a generative model similar to the Vector Quantized-Variational AutoEncoder (VQ-VAE) to acquire prior knowledge for the high-level BEV semantics in the tokenized discrete space. Thanks to the obtained BEV tokens accompanied with a codebook embedding encapsulating the semantics for different BEV elements in the groundtruth maps, we are able to directly align the sparse backbone image features with the obtained BEV tokens from the discrete representation learning based on a specialized token decoder module, and finally generate high-quality BEV maps with the BEV codebook embedding serving as a bridge between PV and BEV. We evaluate the BEV map layout estimation performance of our model, termed VQ-Map, on both the nuScenes and Argoverse benchmarks, achieving 62.2/47.6 mean IoU for surround-view/monocular evaluation on nuScenes, as well as 73.4 IoU for monocular evaluation on Argoverse, which all set a new record for this map layout estimation task. The code and models are available on \url{https://github.com/Z1zyw/VQ-Map}.

cs.CV↗

arXiv2024

Topological entropy and Hausdorff dimension of shrinking target sets

In this paper, we study the topological entropy and the Hausdorff dimension of a shrinking target set. We give lower and upper bounds of topological entropy and Hausdorff dimension for dynamical systems with exponential specification property and Lipschitz continuity for maps and homeomorphisms. It generally applies to uniformly hyperbolic systems, expanding systems, and some symbolic dynamics. We show that lower and upper bounds coincide for both topological entropy and Hausdorff dimension when the systems are hyperbolic automorphisms of torus induced from a matrix with only two different eigenvalues, expanding endomorphism of the torus induced from a matrix with only one eigenvalue or some symbolic systems including one or two-sided shifts of finite type and sofic shifts.

math.DS↗

arXiv2024

Uncovering Attacks and Defenses in Secure Aggregation for Federated Deep Learning

Federated learning enables the collaborative learning of a global model on diverse data, preserving data locality and eliminating the need to transfer user data to a central server. However, data privacy remains vulnerable, as attacks can target user training data by exploiting the updates sent by users during each learning iteration. Secure aggregation protocols are designed to mask/encrypt user updates and enable a central server to aggregate the masked information. MicroSecAgg (PoPETS 2024) proposes a single server secure aggregation protocol that aims to mitigate the high communication complexity of the existing approaches by enabling a one-time setup of the secret to be re-used in multiple training iterations. In this paper, we identify a security flaw in the MicroSecAgg that undermines its privacy guarantees. We detail the security flaw and our attack, demonstrating how an adversary can exploit predictable masking values to compromise user privacy. Our findings highlight the critical need for enhanced security measures in secure aggregation protocols, particularly the implementation of dynamic and unpredictable masking strategies. We propose potential countermeasures to mitigate these vulnerabilities and ensure robust privacy protection in the secure aggregation frameworks.

cs.CR↗

arXiv2024

Graphs of continuous but non-affine functions are never self-similar

Bandt and Kravchenko \cite{BandtKravchenko2010} proved that if a self-similar set spans $\R^m$, then there is no tangent hyperplane at any point of the set. In particular, this indicates that a smooth planar curve is self-similar if and only if it is a straight line. When restricting curves to graphs of continuous functions, we can show that the graph of a continuous function is self-similar if and only if the graph is a straight line, i.e., the underlying function is affine.

math.DS↗

arXiv2024

Fractal geometry of continued fractions with large coefficients and dimension drop problems

In 1928, Jarn\'ık \cite{Jar} obtained that the set of continued fractions with bounded coefficients has Hausdorff dimension one. Good \cite{Goo} observed a dimension drop phenomenon by proving that the Hausdorff dimension of the set of continued fractions whose coefficients tend to infinity is one-half. For the set of continued fractions whose coefficients tend to infinity rapidly, Luczak \cite{Luc} and Feng et al. \cite{FWLT} showed that its Hausdorff dimension decreases even further. Recently, Liao and Rams \cite{LR16} also observed an analogous dimension drop phenomenon when they studied the subexponential growth rate of the sum of coefficients. In this paper, we consolidate and considerably extend the studies of the abovementioned problem into a general dimension drop problem on the distribution of continued fractions with large coefficients. As applications, we use a different approach to reprove a result of Wang and Wu on the dimensions of the Borel-Bernstein sets \cite{WW}, fulfil the dimension gap proposed by Liao and Rams \cite{LR16}, and establish several new results concerning the dimension theory of liminf and limsup sets related to the maximum of coefficients.

math.NT↗

arXiv2024

Safeguarding Vision-Language Models Against Patched Visual Prompt Injectors

Large language models have become increasingly prominent, also signaling a shift towards multimodality as the next frontier in artificial intelligence, where their embeddings are harnessed as prompts to generate textual content. Vision-language models (VLMs) stand at the forefront of this advancement, offering innovative ways to combine visual and textual data for enhanced understanding and interaction. However, this integration also enlarges the attack surface. Patch-based adversarial attack is considered the most realistic threat model in physical vision applications, as demonstrated in many existing literature. In this paper, we propose to address patched visual prompt injection, where adversaries exploit adversarial patches to generate target content in VLMs. Our investigation reveals that patched adversarial prompts exhibit sensitivity to pixel-wise randomization, a trait that remains robust even against adaptive attacks designed to counteract such defenses. Leveraging this insight, we introduce SmoothVLM, a defense mechanism rooted in smoothing techniques, specifically tailored to protect VLMs from the threat of patched visual prompt injectors. Our framework significantly lowers the attack success rate to a range between 0% and 5.0% on two leading VLMs, while achieving around 67.3% to 95.0% context recovery of the benign images, demonstrating a balance between security and usability.

cs.CV↗

arXiv2024

BEV$^2$PR: BEV-Enhanced Visual Place Recognition with Structural Cues

In this paper, we propose a new image-based visual place recognition (VPR) framework by exploiting the structural cues in bird's-eye view (BEV) from a single monocular camera. The motivation arises from two key observations about place recognition methods based on both appearance and structure: 1) For the methods relying on LiDAR sensors, the integration of LiDAR in robotic systems has led to increased expenses, while the alignment of data between different sensors is also a major challenge. 2) Other image-/camera-based methods, involving integrating RGB images and their derived variants (eg, pseudo depth images, pseudo 3D point clouds), exhibit several limitations, such as the failure to effectively exploit the explicit spatial relationships between different objects. To tackle the above issues, we design a new BEV-enhanced VPR framework, namely BEV$^2$PR, generating a composite descriptor with both visual cues and spatial awareness based on a single camera. The key points lie in: 1) We use BEV features as an explicit source of structural knowledge in constructing global features. 2) The lower layers of the pre-trained backbone from BEV generation are shared for visual and structural streams in VPR, facilitating the learning of fine-grained local features in the visual stream. 3) The complementary visual and structural features can jointly enhance VPR performance. Our BEV$^2$PR framework enables consistent performance improvements over several popular aggregation modules for RGB global features. The experiments on our collected VPR-NuScenes dataset demonstrate an absolute gain of 2.47% on Recall@1 for the strong Conv-AP baseline to achieve the best performance in our setting, and notably, a 18.06% gain on the hard set. The code and dataset will be available at https://github.com/FudongGe/BEV2PR.

cs.CV↗