arXiv subjects

Dan He

Publications and source records attributed to Dan He.

10 recordsLinked to original sources

arXiv2025

CroTad: A Contrastive Reinforcement Learning Framework for Online Trajectory Anomaly Detection

Detecting trajectory anomalies is a vital task in modern Intelligent Transportation Systems (ITS), enabling the identification of unsafe, inefficient, or irregular travel behaviours. While deep learning has emerged as the dominant approach, several key challenges remain unresolved. First, sub-trajectory anomaly detection, capable of pinpointing the precise segments where anomalies occur, remains underexplored compared to whole-trajectory analysis. Second, many existing methods depend on carefully tuned thresholds, limiting their adaptability in real-world applications. Moreover, the irregular sampling of trajectory data and the presence of noise in training sets further degrade model performance, making it difficult to learn reliable representations of normal routes. To address these challenges, we propose a contrastive reinforcement learning framework for online trajectory anomaly detection, CroTad. Our method is threshold-free and robust to noisy, irregularly sampled data. By incorporating contrastive learning, CroTad learns to extract diverse normal travel patterns for different itineraries and effectively distinguish anomalous behaviours at both sub-trajectory and point levels. The detection module leverages deep reinforcement learning to perform online, real-time anomaly scoring, enabling timely and fine-grained identification of abnormal segments. Extensive experiments on two real-world datasets demonstrate the effectiveness and robustness of our framework across various evaluation scenarios.

cs.LG↗

arXiv2025

DM-FNet: Unified multimodal medical image fusion via diffusion process-trained encoder-decoder

Multimodal medical image fusion (MMIF) extracts the most meaningful information from multiple source images, enabling a more comprehensive and accurate diagnosis. Achieving high-quality fusion results requires a careful balance of brightness, color, contrast, and detail; this ensures that the fused images effectively display relevant anatomical structures and reflect the functional status of the tissues. However, existing MMIF methods have limited capacity to capture detailed features during conventional training and suffer from insufficient cross-modal feature interaction, leading to suboptimal fused image quality. To address these issues, this study proposes a two-stage diffusion model-based fusion network (DM-FNet) to achieve unified MMIF. In Stage I, a diffusion process trains UNet for image reconstruction. UNet captures detailed information through progressive denoising and represents multilevel data, providing a rich set of feature representations for the subsequent fusion network. In Stage II, noisy images at various steps are input into the fusion network to enhance the model's feature recognition capability. Three key fusion modules are also integrated to process medical images from different modalities adaptively. Ultimately, the robust network structure and a hybrid loss function are integrated to harmonize the fused image's brightness, color, contrast, and detail, enhancing its quality and information density. The experimental results across various medical image types demonstrate that the proposed method performs exceptionally well regarding objective evaluation metrics. The fused image preserves appropriate brightness, a comprehensive distribution of radioactive tracers, rich textures, and clear edges. The code is available at https://github.com/HeDan-11/DM-FNet.

cs.CV↗

arXiv2024

Rethinking Normalization Strategies and Convolutional Kernels for Multimodal Image Fusion

Multimodal image fusion (MMIF) integrates information from different modalities to obtain a comprehensive image, aiding downstream tasks. However, existing research focuses on complementary information fusion and training strategies, overlooking the critical role of underlying architectural components like normalization and convolution kernels. We reevaluate the UNet architecture for end-to-end MMIF, identifying that widely used batch normalization limits performance by smoothing crucial sparse features. To address this, we propose a hybrid of instance and group normalization to maintain sample independence and reinforce intrinsic feature correlations. Crucially, this strategy facilitates richer feature maps, enabling large kernel convolution to fully leverage its receptive field, enhancing detail preservation. Furthermore, the proposed multi-path adaptive fusion module dynamically calibrates features from varying scales and receptive fields, ensuring effective information transfer. Our method achieves SOTA objective performance on MSRS, M$^3$FD, TNO, and Harvard datasets, producing visually clearer salient objects and lesion areas. Notably, it improves MSRS segmentation mIoU by 8.1\% over the infrared image. This performance stems from a synergistic design of normalization and convolution kernels, which preserves critical sparse features. The code is available at https://github.com/HeDan-11/LKC-FUNet.

cs.CV↗

arXiv2022

Pair production of neutral Higgs particles in the B-LSSM

Higgs pair production provides a unique handle for measuring the strength of Higgs self interaction and constraining the shape of the Higgs potential. Including radiative corrections to the trilinear couplings of $CP$-even Higgs, we investigate the cross section of the lightest neutral Higgs pair production in gluon fusion at the Large Hadron Collider in the supersymmetric extensions of the standard model. Numerical results indicate that the correction to the cross section is about 11\% in the B-LSSM, while is only about 4\% in the MSSM. Considering the constraints of the experimental data of the lightest Higgs, we find that the gauge couplings of $U(1)_{B-L}$ and the ratio of the nonzero vacuum expectation values of two singlets also affect strongly the theoretical evaluations on the production cross section in the B-LSSM.

hep-ph↗

arXiv2022

The study of lepton EDMs in $U(1)_X$ SSM

The minimal supersymmetric extension of the standard model (MSSM) is extended to the $U(1)_X$SSM, whose local gauge group is $SU(3)_C \times SU(2)_L \times U(1)_Y \times U(1)_X$. To obtain the $U(1)_X$SSM, we add the new superfields to the MSSM, namely: three Higgs singlets $\hat{\eta},~\hat{\bar{\eta}},~\hat{S}$ and right-handed neutrinos $\hat{\nu}_i$. The CP violating effects are considered to study the lepton electric dipole moment(EDM) in $U(1)_X$SSM. The CP violating phases in $U(1)_X$SSM are more than those in the standard model(SM). In this model, some new parameters $(\theta_S, \theta_{BB^{\prime}}, \theta_{BL})$ as CP violating phases are considered, so there are new contributions to lepton EDMs. It is conducive to exploring the source of CP violation and probing new physical beyond SM.

hep-ph↗

arXiv2020

A Bilateral Game Approach for Task Outsourcing in Multi-access Edge Computing

Multi-access edge computing (MEC) is a promising architecture to provide low-latency applications for future Internet of Things (IoT)-based network systems. Together with the increasing scholarly attention on task offloading, the problem of edge servers' resource allocation has been widely studied. Most of previous works focus on a single edge server (ES) serving multiple terminal entities (TEs), which restricts their access to sufficient resources. In this paper, we consider a MEC resource transaction market with multiple ESs and multiple TEs, which are interdependent and mutually influence each other. However, this many-to-many interaction requires resolving several problems, including task allocation, TEs' selection on ESs and conflicting interests of both parties. Game theory can be used as an effective tool to realize the interests of two or more conflicting individuals in the trading market. Therefore, we propose a bilateral game framework among multiple ESs and multiple TEs by modeling the task outsourcing problem as two noncooperative games: the supplier and customer side games. In the first game, the supply function bidding mechanism is employed to model the ESs' profit maximization problem. The ESs submit their bids to the scheduler, where the computing service price is computed and sent to the TEs. While in the second game, TEs determine the optimal demand profiles according to ESs' bids to maximize their payoff. The existence and uniqueness of the Nash equilibrium in the aforementioned games are proved. A distributed task outsourcing algorithm (DTOA) is designed to determine the equilibrium. Simulation results have demonstrated the superior performance of DTOA in increasing the ESs' profit and TEs' payoff, as well as flattening the peak and off-peak load.

cs.DC↗

arXiv2019

On $t$-relaxed 2-distant circular coloring of graphs

Let $k$ be an positive integer. For any two integers $i$ and $j$ in $\{0,1,\dots,k-1\}$, let $|i-j|_k=\min\{|i-j|,k-|i-j|\}$ be the circular distance between $i$ and $j$. Let $t$ be a nonnegative integer. Suppose $f$ is a mapping from $V(G)$ to $\{0,1,\dots,k-1\}$. If adjacent vertices receive different integers, and for each vertex $u$ of $G$, the number of neighbors $v$ of $u$ with $|f(u)-f(v)|_k=1$ is at most $t$, then $f$ is called a $t$-relaxed 2-distant circular $k$-coloring, or simply a $(\frac{k}{2},t)^*$-coloring of $G$. If $G$ has a $(\frac{k}{2},t)^*$-coloring, then $G$ is called $(\frac{k}{2},t)^*$-colorable. In this paper, we prove that, for any two fixed integers $k$ and $t$ with $k\geq2$ and $t\geq1$, deciding whether $G$ is $(\frac{k}{2},t)^*$-colorable is NP-complete expect the case $k=2$ and the case $k=3$ and $t\leq3$, which are polynomially solvable. For any outerplanar graph $G$, e show that all outerplanar graphs are $(\frac{5}{2},4)^*$-colorable, we prove that there is no fixed positive integer $t$ such that all outerplanar graphs are $(\frac{4}{2},t)^*$-colorable.

math.CO↗

arXiv2017

Arbitrarily Varying Wiretap Channel with State Sequence Known or Unknown at the Receiver

The secrecy capacity problems over the general arbitrarily varying wiretap channel (AVWC), with respect to the maximal decoding error probability and strong secrecy criterion, are considered, where the channel state sequence may be known or unknown at the receiver. In the mean time, it is always assumed that the channel state sequence is known at the eavesdropper and unknown at the transmitter. Capacity results of both stochastic code (with random encoder and deterministic decoder) and random code (with random encoder and decoder) are discussed. This model includes the previous models of classic AVWC as special cases. Single-letter lower bounds on the secrecy capacities are given, which are proved to be the secrecy capacities when the main channel is less noisy than the wiretap channel. The coding scheme is based on Csiszar's almost independent coloring scheme and Ahlswede's elimination technique. Moreover, a new kind of typical sequence with respect to states is defined for this coding scheme. It is concluded that the secrecy capacity of stochastic code is identical to that of random code when the receiver knows the state sequence. Meanwhile, random code may achieve larger secrecy capacity when the state sequence is unknown by the receiver.

cs.IT↗

arXiv2014

IPED2: Inheritance Path based Pedigree Reconstruction Algorithm for Complicated Pedigrees

Reconstruction of family trees, or pedigree reconstruction, for a group of individuals is a fundamental problem in genetics. The problem is known to be NP-hard even for datasets known to only contain siblings. Some recent methods have been developed to accurately and efficiently reconstruct pedigrees. These methods, however, still consider relatively simple pedigrees, for example, they are not able to handle half-sibling situations where a pair of individuals only share one parent. In this work, we propose an efficient method, IPED2, based on our previous work, which specifically targets reconstruction of complicated pedigrees that include half-siblings. We note that the presence of half-siblings makes the reconstruction problem significantly more challenging which is why previous methods exclude the possibility of half-siblings. We proposed a novel model as well as an efficient graph algorithm and experiments show that our algorithm achieves relatively accurate reconstruction. To our knowledge, this is the first method that is able to handle pedigree reconstruction based on genotype data only when half-sibling exists in any generation of the pedigree.

cs.DS↗

arXiv2013

MINT: Mutual Information based Transductive Feature Selection for Genetic Trait Prediction

Whole genome prediction of complex phenotypic traits using high-density genotyping arrays has attracted a great deal of attention, as it is relevant to the fields of plant and animal breeding and genetic epidemiology. As the number of genotypes is generally much bigger than the number of samples, predictive models suffer from the curse-of-dimensionality. The curse-of-dimensionality problem not only affects the computational efficiency of a particular genomic selection method, but can also lead to poor performance, mainly due to correlation among markers. In this work we proposed the first transductive feature selection method based on the MRMR (Max-Relevance and Min-Redundancy) criterion which we call MINT. We applied MINT on genetic trait prediction problems and showed that in general MINT is a better feature selection method than the state-of-the-art inductive method mRMR.

cs.LG↗