Source author record

Tiantian Li

Tiantian Li appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computer Vision cond-mat.quant-gas Cryptography and Security Machine Learning Artificial Intelligence eess.IV math.PR nlin.SI physics.optics q-fin.PM quant-ph Robotics

Catalog footprint

What is connected

13works

12topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Dynamic-TD3: A Novel Algorithm for UAV Path Planning with Dynamic Obstacle Trajectory Prediction

Deep reinforcement learning (DRL) finds extensive application in autonomous drone navigation within complex, high-risk environments. However, its practical deployment faces a safety-exploration dilemma: soft penalty mechanisms encourage risky trial-and-error, while most constraint-based methods suffer degraded performance under sensor noise and intent uncertainty. We propose Dynamic-TD3, a physically enhanced framework that enforces strict safety constraints while maintaining maneuverability by modeling navigation as a Constrained Markov Decision Process (CMDP). This framework integrates an Adaptive Trajectory Relational Evolution Mechanism (ATREM) to capture long-range intentions and employs a Physically Aware Gated Kalman Filter (PAG-KF) to mitigate non-stationary observation noise. The resulting state representation drives a dual-criterion policy that balances mission efficiency against hard safety constraints via Lagrangian relaxation. In experiments with aggressive dynamic threats, this approach demonstrates superior collision avoidance performance, reduced energy consumption, and smoother flight trajectories.

preprint2025arXiv

GaussianImage++: Boosted Image Representation and Compression with 2D Gaussian Splatting

Implicit neural representations (INRs) have achieved remarkable success in image representation and compression, but they require substantial training time and memory. Meanwhile, recent 2D Gaussian Splatting (GS) methods (\textit{e.g.}, GaussianImage) offer promising alternatives through efficient primitive-based rendering. However, these methods require excessive Gaussian primitives to maintain high visual fidelity. To exploit the potential of GS-based approaches, we present GaussianImage++, which utilizes limited Gaussian primitives to achieve impressive representation and compression performance. Firstly, we introduce a distortion-driven densification mechanism. It progressively allocates Gaussian primitives according to signal intensity. Secondly, we employ context-aware Gaussian filters for each primitive, which assist in the densification to optimize Gaussian primitives based on varying image content. Thirdly, we integrate attribute-separated learnable scalar quantizers and quantization-aware training, enabling efficient compression of primitive attributes. Experimental results demonstrate the effectiveness of our method. In particular, GaussianImage++ outperforms GaussianImage and INRs-based COIN in representation and compression performance while maintaining real-time decoding and low memory usage.

preprint2022arXiv

BABD: A Bitcoin Address Behavior Dataset for Pattern Analysis

Cryptocurrencies are no longer just the preferred option for cybercriminal activities on darknets, due to the increasing adoption in mainstream applications. This is partly due to the transparency associated with the underpinning ledgers, where any individual can access the record of a transaction record on the public ledger. In this paper, we build a dataset comprising Bitcoin transactions between 12 July 2019 and 26 May 2021. This dataset (hereafter referred to as BABD-13) contains 13 types of Bitcoin addresses, 5 categories of indicators with 148 features, and 544,462 labeled data, which is the largest labeled Bitcoin address behavior dataset publicly available to our knowledge. We then use our proposed dataset on common machine learning models, namely: k-nearest neighbors algorithm, decision tree, random forest, multilayer perceptron, and XGBoost. The results show that the accuracy rates of these machine learning models for the multi-classification task on our proposed dataset are between 93.24% and 97.13%. We also analyze the proposed features and their relationships from the experiments, and propose a k-hop subgraph generation algorithm to extract a k-hop subgraph from the entire Bitcoin transaction graph constructed by the directed heterogeneous multigraph starting from a specific Bitcoin address node (e.g., a known transaction associated with a criminal investigation). Besides, we initially analyze the behavior patterns of different types of Bitcoin addresses according to the extracted features.

preprint2022arXiv

Generating Image Adversarial Examples by Embedding Digital Watermarks

With the increasing attention to deep neural network (DNN) models, attacks are also upcoming for such models. For example, an attacker may carefully construct images in specific ways (also referred to as adversarial examples) aiming to mislead the DNN models to output incorrect classification results. Similarly, many efforts are proposed to detect and mitigate adversarial examples, usually for certain dedicated attacks. In this paper, we propose a novel digital watermark-based method to generate image adversarial examples to fool DNN models. Specifically, partial main features of the watermark image are embedded into the host image almost invisibly, aiming to tamper with and damage the recognition capabilities of the DNN models. We devise an efficient mechanism to select host images and watermark images and utilize the improved discrete wavelet transform (DWT) based Patchwork watermarking algorithm with a set of valid hyperparameters to embed digital watermarks from the watermark image dataset into original images for generating image adversarial examples. The experimental results illustrate that the attack success rate on common DNN models can reach an average of 95.47% on the CIFAR-10 dataset and the highest at 98.71%. Besides, our scheme is able to generate a large number of adversarial examples efficiently, concretely, an average of 1.17 seconds for completing the attacks on each image on the CIFAR-10 dataset. In addition, we design a baseline experiment using the watermark images generated by Gaussian noise as the watermark image dataset that also displays the effectiveness of our scheme. Similarly, we also propose the modified discrete cosine transform (DCT) based Patchwork watermarking algorithm. To ensure repeatability and reproducibility, the source code is available on GitHub.

preprint2021arXiv

A Lightweight Privacy-Preserving Scheme Using Label-based Pixel Block Mixing for Image Classification in Deep Learning

To ensure the privacy of sensitive data used in the training of deep learning models, a number of privacy-preserving methods have been designed by the research community. However, existing schemes are generally designed to work with textual data, or are not efficient when a large number of images is used for training. Hence, in this paper we propose a lightweight and efficient approach to preserve image privacy while maintaining the availability of the training set. Specifically, we design the pixel block mixing algorithm for image classification privacy preservation in deep learning. To evaluate its utility, we use the mixed training set to train the ResNet50, VGG16, InceptionV3 and DenseNet121 models on the WIKI dataset and the CNBC face dataset. Experimental findings on the testing set show that our scheme preserves image privacy while maintaining the availability of the training set in the deep learning models. Additionally, the experimental results demonstrate that we achieve good performance for the VGG16 model on the WIKI dataset and both ResNet50 and DenseNet121 on the CNBC dataset. The pixel block algorithm achieves fairly high efficiency in the mixing of the images, and it is computationally challenging for the attackers to restore the mixed training set to the original training set. Moreover, data augmentation can be applied to the mixed training set to improve the training's effectiveness.

preprint2020arXiv

Deep Complementary Joint Model for Complex Scene Registration and Few-shot Segmentation on Medical Images

Deep learning-based medical image registration and segmentation joint models utilize the complementarity (augmentation data or weakly supervised data from registration, region constraints from segmentation) to bring mutual improvement in complex scene and few-shot situation. However, further adoption of the joint models are hindered: 1) the diversity of augmentation data is reduced limiting the further enhancement of segmentation, 2) misaligned regions in weakly supervised data disturb the training process, 3) lack of label-based region constraints in few-shot situation limits the registration performance. We propose a novel Deep Complementary Joint Model (DeepRS) for complex scene registration and few-shot segmentation. We embed a perturbation factor in the registration to increase the activity of deformation thus maintaining the augmentation data diversity. We take a pixel-wise discriminator to extract alignment confidence maps which highlight aligned regions in weakly supervised data so the misaligned regions' disturbance will be suppressed via weighting. The outputs from segmentation model are utilized to implement deep-based region constraints thus relieving the label requirements and bringing fine registration. Extensive experiments on the CT dataset of MM-WHS 2017 Challenge show great advantages of our DeepRS that outperforms the existing state-of-the-art models.

preprint2020arXiv

Semi-supervised learning method based on predefined evenly-distributed class centroids

Compared to supervised learning, semi-supervised learning reduces the dependence of deep learning on a large number of labeled samples. In this work, we use a small number of labeled samples and perform data augmentation on unlabeled samples to achieve image classification. Our method constrains all samples to the predefined evenly-distributed class centroids (PEDCC) by the corresponding loss function. Specifically, the PEDCC-Loss for labeled samples, and the maximum mean discrepancy loss for unlabeled samples are used to make the feature distribution closer to the distribution of PEDCC. Our method ensures that the inter-class distance is large and the intra-class distance is small enough to make the classification boundaries between different classes clearer. Meanwhile, for unlabeled samples, we also use KL divergence to constrain the consistency of the network predictions between unlabeled and augmented samples. Our semi-supervised learning method achieves the state-of-the-art results, with 4000 labeled samples on CIFAR10 and 1000 labeled samples on SVHN, and the accuracy is 95.10% and 97.58% respectively.

preprint2016arXiv

Dynamics of a coupled spin vortex pair in dipolar spinor Bose-Einstein condensates

The collisional and magnetic field quench dynamics of a coupled spin-vortex pair in dipolar spinor Bose-Einstein condensates in a double well potential are numerically investigated in the mean field theory. Upon a sudden release of the potential barrier the two layers of condensates collide with each other in the trap center with the chirality of the vortex pair exchanged after each collision, showing the typical signature of in-phase collision for the parallel spin vortex phase, and out-of-phase collision for the antiparallel phase. When quenching the transverse magnetic field, the vortex center in the single-layered condensate starts to make a helical motion with oval-shaped trajectories and the displacement of the center position is found to exhibit a damped simple harmonic oscillation with an intrinsic frequency and damping rate. The oscillation mode of the spin vortex pair may be tuned by the initial magnetic field and the height of the Gaussian barrier, e.g. the gyrotropic motions for parallel spin vortex pair are out of sync with each other in the two layers, while those for the antiparallel pair exhibit a double-helix-structure with the vortex centers moving opposite to each other with the same amplitude.

preprint2015arXiv

Coupled spin-vortex pair in dipolar spinor Bose-Einstein condensates

We investigate the ground-state and magnetic properties of a dipolar spin-1 Bose-Einstein condensate trapped in a symmetric double-well potential. In particular, we focus on the spin-vortex states by assuming that each potential well is highly pancake shaped. We show that the presence of the double-well potential gives rise to two different spin configurations for the spin-vortex pair states. We also study the response of the coupled spin-vortex pair to static transverse magnetic fields.

preprint2015arXiv

Symmetry protected skyrmions in 3D spin-orbit coupled Bose gases

We present a variational study of pseudo-spin $1/2$ Bose gases in a harmonic trap with weak 3D spin-orbit coupling of $\bmsigma\cdot\mathbf{p}$ type. This spin-orbit coupling mixes states with different parities, which inspires us to approximate the single particle state with the eigenstates of the total angular momentum, i.e. superposition of harmonic $s$-wave and $p$-wave states. As the time reversal symmetry is protected by two-body interaction, we set the variational order parameter as the combination of two mutually time reversal symmetric eigenstates of the total angular momentum. The variational results essentially reproduce the 3D skyrmion-like ground state recently identified by Kawakami {\it et al.}. We show that these skyrmion-like ground states emerging in this model are primarily caused by $p$ wave spatial mode involving in the variational order parameter that drives two spin components spatially separated. We find the ground state of this system falls into two phases with different density distribution symmetries depending on the relative magnitude of intraspecies and interspecies interaction: Phase I has parity symmetric and axisymmetric density distributions, while Phase II is featured with special joint symmetries of discrete rotational and time reversal symmetry. With the increasing interaction strength the transition occurs between two phases with distinct density distributions, while the topological 3D skyrmion-like spin texture is symmetry protected.

preprint2013arXiv

Darboux transformation and solutions of the two-component Hirota-Maxwell-Bloch system

In this paper, we derive n-fold Darboux transformation of the two-component Hirota and the Maxwell-Bloch(TH-MB) equations and its determinant representation. Using Darboux determinant representation, we provide soliton solutions, positon solutions of the TH-MB equations.

preprint2011arXiv

Interspecies singlet pairing in a mixture of two spin-1 Bose condensates

We study the ground state properties of a mixture formed by two spin-1 condensates in the absence of an external magnetic field. As the collisional symmetry between interspecies bosonic atoms is broken, the interspecies coupling interaction ($β$) and interspecies singlet pairing interaction ($γ$) arise. The ground state can be calculated using the angular momentum theory analytically for $γ=0$. The full quantum approach of exact diagonalization is adopted numerically to consider the more general case as $γ\neq 0$. We illustrate the competition between the two interspecies interactions and find that as singlet pairing interaction dominates (or the total spin vanishes), there are still different types of singlet formations which are well determined by $β$.

preprint2010arXiv

Optimal mean-variance investment strategy under value-at-risk constraints

This paper is devoted to study the effects arising from imposing a value-at-risk (VaR) constraint in mean-variance portfolio selection problem for an investor who receives a stochastic cash flow which he/she must then invest in a continuous-time financial market. For simplicity, we assume that there is only one investment opportunity available for the investor, a risky stock. Using techniques of stochastic linear-quadratic (LQ) control, the optimal mean-variance investment strategy with and without VaR constraint are derived explicitly in closed forms, based on solution of corresponding Hamilton-Jacobi-Bellman (HJB) equation. Furthermore, some numerical examples are proposed to show how the addition of the VaR constraint affects the optimal strategy.