arXiv subjects

Yanhua Zhao

Publications and source records attributed to Yanhua Zhao.

11 recordsLinked to original sources

arXiv2026

Activity Recognition from Smart Insole Sensor Data Using a Circular Dilated CNN

Smart insoles equipped with pressure sensors, accelerometers, and gyroscopes offer a non-intrusive means of monitoring human gait and posture. We present an activity classification system based on a circular dilated convolutional neural network (CDCNN) that processes multi-modal time-series data from such insoles. The model operates on 160-frame windows with 24 channels (18 pressure, 3 accelerometer, 3 gyroscope axes), achieving 86.42% test accuracy in a subject-independent evaluation on a four-class task (Standing, Walking, Sitting, Tandem), compared with 87.83% for an extreme gradient-boosted tree (XGBoost) model trained on flattened data. Permutation feature importance reveals that inertial sensors (accelerometer and gyroscope) contribute substantially to discrimination. The approach is suitable for embedded deployment and real-time inference.

cs.LG↗

arXiv2026

An Example for Domain Adaptation Using CycleGAN

Cycle-Consistent Adversarial Network (CycleGAN) is very promising in domain adaptation. In this report, an example in medical domain will be explained. We present struecture of a CycleGAN model for unpaired image-to-image translation from microscopy to pseudo H\&E stained histopathology images.

cs.CV↗

arXiv2026

Attention Consistency Regularization for Interpretable Early-Exit Neural Networks

Early-exit neural networks enable adaptive inference by allowing predictions at intermediate layers, reducing computational cost. However, early exits often lack interpretability and may focus on different features than deeper layers, limiting trust and explainability. This paper presents Explanation-Guided Training (EGT), a multi-objective framework that improves interpretability and consistency in early-exit networks through attention-based regularization. EGT introduces an attention consistency loss that aligns early-exit attention maps with the final exit. The framework jointly optimizes classification accuracy and attention consistency through a weighted combination of losses. Experiments on a real-world image classification dataset demonstrate that EGT achieves up to 98.97% overall accuracy (matching baseline performance) with a 1.97x inference speedup through early exits, while improving attention consistency by up to 18.5% compared to baseline models. The proposed method provides more interpretable and consistent explanations across all exit points, making early-exit networks more suitable for explainable AI applications in resource-constrained environments.

cs.LG↗

arXiv2026

Online parameter estimation for the Crazyflie quadcopter through an EM algorithm

Drones are becoming more and more popular nowadays. They are small in size, low in cost, and reliable in operation. They contain a variety of sensors and can perform a variety of flight tasks, reaching places that are difficult or inaccessible for humans. Earthquakes damage a lot of infrastructure, making it impossible for rescuers to reach some areas. But drones can help. Many amateur and professional photographers like to use drones for aerial photography. Drones play a non-negligible role in agriculture and transportation too. Drones can be used to spray pesticides, and they can also transport supplies. A quadcopter is a four-rotor drone and has been studied in this paper. In this paper, random noise is added to the quadcopter system and its effects on the drone system are studied. An extended Kalman filter has been used to estimate the state based on noisy observations from the sensor. Based on a SDE system, a linear quadratic Gaussian controller has been implemented. The expectation maximization algorithm has been applied for parameter estimation of the quadcopter. The results of offline parameter estimation and online parameter estimation are presented. The results show that the online parameter estimation has a slightly larger range of convergence values than the offline parameter estimation.

cs.AI↗

arXiv2023

Spectral extremal problem on the square of $\ell$-cycle

Let $C_{\ell}$ be the cycle of order ${\ell}$. The square of $C_{\ell}$, denoted by $C_{\ell}^2$, is obtained by joining all pairs of vertices with distance no more than two in $C_{\ell}$. A graph is called $F$-free if it does not contain $F$ as a subgraph. Denote by $ex(n,F)$ and $spex(n,F)$ the maximum size and spectral radius over all $n$-vertex $F$-free graphs, respectively. The well-known Turán problem asks for the $ex(n,F)$, and Nikiforov in 2010 proposed a spectral counterpart, known as Brualdi-Solheid-Turán type problem, focusing on determining $spex(n,F)$. In this paper, we consider a Turán problem on $ex(n,C_{\ell}^2)$ and a Brualdi--Solheid--Turán type problem on $spex(n,C_{\ell}^2)$. We give a sharp bound of $ex(n,C_{\ell}^2)$ and $spex(n,C_{\ell}^2)$ for sufficiently large $n$, respectively. Moreover, in both results, we characterize the corresponding extremal graphs for any integer $\ell\geq 6$ that is not divisible by $3$.

math.CO↗

arXiv2022

A spectral condition for the existence of cycles with consecutive odd lengths in non-bipartite graphs

A graph $G$ is called $H$-free, if it does not contain $H$ as a subgraph. In 2010, Nikiforov proposed a Brualdi-Solheid-Turán type problem: what is the maximum spectral radius of an $H$-free graph of order $n$? In this paper, we consider the Brualdi-Solheid-Turán type problem for non-bipartite graphs. Let $K_{a, b}\bullet K_3$ denote the graph obtained by identifying a vertex of $K_{a,b}$ in the part of size $b$ and a vertex of $K_3$. We prove that if $G$ is a non-bipartite graph of order $n$ satisfying $ρ(G)\geq ρ(K_{\lceil\frac{n-2}{2}\rceil, \lfloor\frac{n-2}{2}\rfloor}\bullet K_3)$, then $G$ contains all odd cycles $C_{2l+1}$ for each integer $l\in[2,k]$ unless $G\cong K_{\lceil\frac{n-2}{2}\rceil, \lfloor\frac{n-2}{2}\rfloor}\bullet K_3$, provided that $n$ is sufficiently large with respect to $k$. This resolves the problem posed by Guo, Lin and Zhao (2021).

math.CO↗

arXiv2021

Spectral radius, edge-disjoint cycles and cycles of the same length

In this paper, we give spectral conditions to guarantee the existence of two edge disjoint cycles and two cycles of the same length. These two results can be seen as spectral analogues of Erdős and Posa's size condition and Erdős' classic problem on non existence of two cycles of the same length. By using double leading eigenvectors skill, we further give spectral condition to guarantee the existence of $k$ edge disjoint triangles.

math.CO↗

arXiv2021

The $A_α$-spectral radius and perfect matchings of graphs

Let $α\in[0,1)$, and let $G$ be a graph of even order $n$ with $n\geq f(α)$, where $f(α)=10$ for $0\leq α\leq1/2$, $f(α)=14$ for $1/2<α\leq 2/3$ and $f(α)=5/(1-α)$ for $2/3<α<1$. In this paper, it is shown that if the $A_α$-spectral radius of $G$ is not less than the largest root of $x^3 - ((α+ 1)n +α-4)x^2 + (αn^2 + (α^2 - 2α- 1)n - 2α+1)x -α^2n^2 + (5α^2 - 3α+ 2)n - 10α^2 + 15α- 8=0$ then $G$ has a perfect matching unless $G=K_1\nabla(K_{n-3}\cup 2K_1)$. This generalizes a result of S. O [Spectral radius and matchings in graphs, Linear Algebra Appl. 614 (2021) 316--324], which gives a sufficient condition for the existence of a perfect matching in a graph in terms of the adjacency spectral radius.

math.CO↗

arXiv2020

The signless Laplacian spectral radius of graphs with no intersecting triangles

Let $F_k$ denote the $k$-fan consisting of $k$ triangles which intersect in exactly one common vertex, and $S_{n,k}$ the complete split graph of order $n$ consisting of a clique on $k$ vertices and an independent set on the remaining vertices in which each vertex of the clique is adjacent to each vertex of the independent set. In this paper, it is shown that $S_{n,k}$ is the unique graph attaining the maximum signless Laplacian spectral radius among all graphs of order $n$ containing no $F_k$, provided that $k\geq 2$ and $n\geq 3k^2-k-2$.

math.CO↗

arXiv2020

The maximum spectral radius of wheel-free graphs

A wheel graph is a graph formed by connecting a single vertex to all vertices of a cycle. A graph is called wheel-free if it does not contain any wheel graph as a subgraph. In 2010, Nikiforov proposed a Brualdi-Solheid-Turán type problem: what is the maximum spectral radius of a graph of order $n$ that does not contain subgraphs of particular kind. In this paper, we study the Brualdi-Solheid-Turán type problem for wheel-free graphs, and we determine the maximum (signless Laplacian) spectral radius of a wheel-free graph of order $n$. Furthermore, we characterize the extremal graphs.

math.CO↗

arXiv2015

Bioinspired interfacial materials with enhanced drop mobility: From fundamentals to multifunctional applications

The development of bio-inspired interfacial materials with enhanced drop mobility that mimic the innate functionalities of nature will have significant impact on the energy, environment and global healthcare. In spite of extensive progress, the state of the art of interfacial materials have not reached the level of maturity sufficient for industrial applications in terms of scalability, stability and reliability, which are complicated by their operating environments and lack of facile approaches to exquisitely control the local structural texture and chemical composition at multiple length scales. In this review, we focus on the recent advances in the fundamental understanding as well as practical applications of bio-inspired interfacial materials, with an emphasis on the drop impact induced bouncing and coalescence induced jumping behaviors. We also suggest our own perspectives on how to catalyze new discoveries and to foster technological adoption to move this exciting area forward.

cond-mat.mtrl-sci↗