arXiv subjects

Lin Li

Publications and source records attributed to Lin Li.

At least 253 records · Page 14Linked to original sources

arXiv2023

Cross-domain Robust Deepfake Bias Expansion Network for Face Forgery Detection

The rapid advancement of deepfake technologies raises significant concerns about the security of face recognition systems. While existing methods leverage the clues left by deepfake techniques for face forgery detection, malicious users may intentionally manipulate forged faces to obscure the traces of deepfake clues and thereby deceive detection tools. Meanwhile, attaining cross-domain robustness for data-based methods poses a challenge due to potential gaps in the training data, which may not encompass samples from all relevant domains. Therefore, in this paper, we introduce a solution - a Cross-Domain Robust Bias Expansion Network (BENet) - designed to enhance face forgery detection. BENet employs an auto-encoder to reconstruct input faces, maintaining the invariance of real faces while selectively enhancing the difference between reconstructed fake faces and their original counterparts. This enhanced bias forms a robust foundation upon which dependable forgery detection can be built. To optimize the reconstruction results in BENet, we employ a bias expansion loss infused with contrastive concepts to attain the aforementioned objective. In addition, to further heighten the amplification of forged clues, BENet incorporates a Latent-Space Attention (LSA) module. This LSA module effectively captures variances in latent features between the auto-encoder's encoder and decoder, placing emphasis on inconsistent forgery-related information. Furthermore, BENet incorporates a cross-domain detector with a threshold to determine whether the sample belongs to a known distribution. The correction of classification results through the cross-domain detector enables BENet to defend against unknown deepfake attacks from cross-domain. Extensive experiments demonstrate the superiority of BENet compared with state-of-the-art methods in intra-database and cross-database evaluations.

cs.CV↗

arXiv2023

VisionFM: a Multi-Modal Multi-Task Vision Foundation Model for Generalist Ophthalmic Artificial Intelligence

We present VisionFM, a foundation model pre-trained with 3.4 million ophthalmic images from 560,457 individuals, covering a broad range of ophthalmic diseases, modalities, imaging devices, and demography. After pre-training, VisionFM provides a foundation to foster multiple ophthalmic artificial intelligence (AI) applications, such as disease screening and diagnosis, disease prognosis, subclassification of disease phenotype, and systemic biomarker and disease prediction, with each application enhanced with expert-level intelligence and accuracy. The generalist intelligence of VisionFM outperformed ophthalmologists with basic and intermediate levels in jointly diagnosing 12 common ophthalmic diseases. Evaluated on a new large-scale ophthalmic disease diagnosis benchmark database, as well as a new large-scale segmentation and detection benchmark database, VisionFM outperformed strong baseline deep neural networks. The ophthalmic image representations learned by VisionFM exhibited noteworthy explainability, and demonstrated strong generalizability to new ophthalmic modalities, disease spectrum, and imaging devices. As a foundation model, VisionFM has a large capacity to learn from diverse ophthalmic imaging data and disparate datasets. To be commensurate with this capacity, in addition to the real data used for pre-training, we also generated and leveraged synthetic ophthalmic imaging data. Experimental results revealed that synthetic data that passed visual Turing tests, can also enhance the representation learning capability of VisionFM, leading to substantial performance gains on downstream ophthalmic AI tasks. Beyond the ophthalmic AI applications developed, validated, and demonstrated in this work, substantial further applications can be achieved in an efficient and cost-effective manner using VisionFM as the foundation.

eess.IV↗

arXiv2023

Large AI Models in Health Informatics: Applications, Challenges, and the Future

Large AI models, or foundation models, are models recently emerging with massive scales both parameter-wise and data-wise, the magnitudes of which can reach beyond billions. Once pretrained, large AI models demonstrate impressive performance in various downstream tasks. A prime example is ChatGPT, whose capability has compelled people's imagination about the far-reaching influence that large AI models can have and their potential to transform different domains of our lives. In health informatics, the advent of large AI models has brought new paradigms for the design of methodologies. The scale of multi-modal data in the biomedical and health domain has been ever-expanding especially since the community embraced the era of deep learning, which provides the ground to develop, validate, and advance large AI models for breakthroughs in health-related areas. This article presents a comprehensive review of large AI models, from background to their applications. We identify seven key sectors in which large AI models are applicable and might have substantial influence, including 1) bioinformatics; 2) medical diagnosis; 3) medical imaging; 4) medical informatics; 5) medical education; 6) public health; and 7) medical robotics. We examine their challenges, followed by a critical discussion about potential future directions and pitfalls of large AI models in transforming the field of health informatics.

cs.AI↗

arXiv2023

A Survey of Multimodal Information Fusion for Smart Healthcare: Mapping the Journey from Data to Wisdom

Multimodal medical data fusion has emerged as a transformative approach in smart healthcare, enabling a comprehensive understanding of patient health and personalized treatment plans. In this paper, a journey from data to information to knowledge to wisdom (DIKW) is explored through multimodal fusion for smart healthcare. We present a comprehensive review of multimodal medical data fusion focused on the integration of various data modalities. The review explores different approaches such as feature selection, rule-based systems, machine learning, deep learning, and natural language processing, for fusing and analyzing multimodal data. This paper also highlights the challenges associated with multimodal fusion in healthcare. By synthesizing the reviewed frameworks and theories, it proposes a generic framework for multimodal medical data fusion that aligns with the DIKW model. Moreover, it discusses future directions related to the four pillars of healthcare: Predictive, Preventive, Personalized, and Participatory approaches. The components of the comprehensive survey presented in this paper form the foundation for more successful implementation of multimodal fusion in smart healthcare. Our findings can guide researchers and practitioners in leveraging the power of multimodal fusion with the state-of-the-art approaches to revolutionize healthcare and improve patient outcomes.

cs.IR↗

arXiv2023

PDRL: Multi-Agent based Reinforcement Learning for Predictive Monitoring

Reinforcement learning has been increasingly applied in monitoring applications because of its ability to learn from previous experiences and can make adaptive decisions. However, existing machine learning-based health monitoring applications are mostly supervised learning algorithms, trained on labels and they cannot make adaptive decisions in an uncertain complex environment. This study proposes a novel and generic system, predictive deep reinforcement learning (PDRL) with multiple RL agents in a time series forecasting environment. The proposed generic framework accommodates virtual Deep Q Network (DQN) agents to monitor predicted future states of a complex environment with a well-defined reward policy so that the agent learns existing knowledge while maximizing their rewards. In the evaluation process of the proposed framework, three DRL agents were deployed to monitor a subject's future heart rate, respiration, and temperature predicted using a BiLSTM model. With each iteration, the three agents were able to learn the associated patterns and their cumulative rewards gradually increased. It outperformed the baseline models for all three monitoring agents. The proposed PDRL framework is able to achieve state-of-the-art performance in the time series forecasting process. The proposed DRL agents and deep learning model in the PDRL framework are customized to implement the transfer learning in other forecasting applications like traffic and weather and monitor their states. The PDRL framework is able to learn the future states of the traffic and weather forecasting and the cumulative rewards are gradually increasing over each episode.

cs.LG↗

arXiv2023

Towards Runtime Customizable Trusted Execution Environment on FPGA-SoC

Processing sensitive data and deploying well-designed Intellectual Property (IP) cores on remote Field Programmable Gate Array (FPGA) are prone to private data leakage and IP theft. One effective solution is constructing Trusted Execution Environment (TEE) on FPGA-SoCs (FPGA System on Chips). Researchers have integrated this type TEE with Trusted Platform Module (TPM)-based trusted boot, denoted as FPGA-SoC tbTEE. But there is no effort on secure and trusted runtime customization of FPGA-SoC TEE. This paper extends FPGA-SoC tbTEE to build Runtime Customizable TEE (RCTEE) on FPGA-SoC by additive three major components (our work): 1) CrloadIP, which can load an IP core at runtime such that RCTEE can be adjusted dynamically and securely; 2) CexecIP, which can not only execute an IP core without modifying the operating system of FPGA-SoC TEE, but also prevent insider attacks from executing IPs deployed in RCTEE; 3) CremoAT, which can provide the newly measured RCTEE state and establish a secure and trusted communication path between remote verifiers and RCTEE. We conduct a security analysis of RCTEE and its performance evaluation on Xilinx Zynq UltraScale+ XCZU15EG 2FFVB1156 MPSoC.

cs.CR↗

arXiv2023

Ballistic PbTe Nanowire Devices

Disorder is the primary obstacle in current Majorana nanowire experiments. Reducing disorder or achieving ballistic transport is thus of paramount importance. In clean and ballistic nanowire devices, quantized conductance is expected with plateau quality serving as a benchmark for disorder assessment. Here, we introduce ballistic PbTe nanowire devices grown using the selective-area-growth (SAG) technique. Quantized conductance plateaus in units of $2e^2/h$ are observed at zero magnetic field. This observation represents an advancement in diminishing disorder within SAG nanowires, as none of the previously studied SAG nanowires (InSb or InAs) exhibit zero-field ballistic transport. Notably, the plateau values indicate that the ubiquitous valley degeneracy in PbTe is lifted in nanowire devices. This degeneracy lifting addresses an additional concern in the pursuit of Majorana realization. Moreover, these ballistic PbTe nanowires may enable the search for clean signatures of the spin-orbit helical gap in future devices.

cond-mat.mes-hall↗

arXiv2023

Hard superconducting gap in PbTe nanowires

Semiconductor nanowires coupled to a superconductor provide a powerful testbed for quantum device physics such as Majorana zero modes and gate-tunable hybrid qubits. The performance of these quantum devices heavily relies on the quality of the induced superconducting gap. A hard gap, evident as vanishing subgap conductance in tunneling spectroscopy, is both necessary and desired. Previously, a hard gap has been achieved and extensively studied in III-V semiconductor nanowires (InAs and InSb). In this study, we present the observation of a hard superconducting gap in PbTe nanowires coupled to a superconductor Pb. The gap size ($Δ$) is $\sim$ 1 meV (maximally 1.3 meV in one device). Additionally, subgap Andreev bound states can also be created and controlled through gate tuning. Tuning a device into the open regime can reveal Andreev enhancement of the subgap conductance, suggesting a remarkable transparent superconductor-semiconductor interface, with a transparency of $\sim$ 0.96. These results pave the way for diverse superconducting quantum devices based on PbTe nanowires.

cond-mat.mes-hall↗

arXiv2023

Identifying symbiotic stars with machine learning

Symbiotic stars are interacting binary systems, making them valuable for studying various astronomical phenomena, such as stellar evolution, mass transfer, and accretion processes. Despite recent progress in the discovery of symbiotic stars, a significant discrepancy between the observed population of symbiotic stars and the number predicted by theoretical models. To bridge this gap, this study utilized machine learning techniques to efficiently identify new symbiotic stars candidates. Three algorithms (XGBoost, LightGBM, and Decision Tree) were applied to a dataset of 198 confirmed symbiotic stars and the resulting model was then used to analyze data from the LAMOST survey, leading to the identification of 11,709 potential symbiotic stars candidates. Out of the these potential symbiotic stars candidates listed in the catalog, 15 have spectra available in the SDSS survey. Among these 15 candidates, two candidates, namely V* V603 Ori and V* GN Tau, have been confirmed as symbiotic stars. The remaining 11 candidates have been classified as accreting-only symbiotic star candidates. The other two candidates, one of which has been identified as a galaxy by both SDSS and LAMOST surveys, and the other identified as a quasar by SDSS survey and as a galaxy by LAMOST survey.

astro-ph.SR↗

arXiv2023

Interpretable Style Transfer for Text-to-Speech with ControlVAE and Diffusion Bridge

With the demand for autonomous control and personalized speech generation, the style control and transfer in Text-to-Speech (TTS) is becoming more and more important. In this paper, we propose a new TTS system that can perform style transfer with interpretability and high fidelity. Firstly, we design a TTS system that combines variational autoencoder (VAE) and diffusion refiner to get refined mel-spectrograms. Specifically, a two-stage and a one-stage system are designed respectively, to improve the audio quality and the performance of style transfer. Secondly, a diffusion bridge of quantized VAE is designed to efficiently learn complex discrete style representations and improve the performance of style transfer. To have a better ability of style transfer, we introduce ControlVAE to improve the reconstruction quality and have good interpretability simultaneously. Experiments on LibriTTS dataset demonstrate that our method is more effective than baseline models.

cs.SD↗

arXiv2023

Community Detection Graph Convolutional Network for Overlap-Aware Speaker Diarization

The clustering algorithm plays a crucial role in speaker diarization systems. However, traditional clustering algorithms suffer from the complex distribution of speaker embeddings and lack of digging potential relationships between speakers in a session. We propose a novel graph-based clustering approach called Community Detection Graph Convolutional Network (CDGCN) to improve the performance of the speaker diarization system. The CDGCN-based clustering method consists of graph generation, sub-graph detection, and Graph-based Overlapped Speech Detection (Graph-OSD). Firstly, the graph generation refines the local linkages among speech segments. Secondly the sub-graph detection finds the optimal global partition of the speaker graph. Finally, we view speaker clustering for overlap-aware speaker diarization as an overlapped community detection task and design a Graph-OSD component to output overlap-aware labels. By capturing local and global information, the speaker diarization system with CDGCN clustering outperforms the traditional Clustering-based Speaker Diarization (CSD) systems on the DIHARD III corpus.

eess.AS↗

arXiv2023

Results of 23 yr of Pulsar Timing of PSR J1453-6413

In this paper, we presented the 23.3 years of pulsar timing results of PSR J1456-6413 based on the observation of Parkes 64m radio telescope. We detected two new glitches at MJD 57093(3) and 59060(12) and confirmed its first glitch at MJD 54554(10). Using the "Cholesky" timing analysis method, we have determined its position, proper motion, and two-dimensional transverse velocities from the data segments before and after the second glitch, respectively. Furthermore, we detected exponential recovery behavior after the first glitch, with a recovery time scale of approximately 200 days and a corresponding exponential recovery factor Q of approximately 0.15(2), while no exponential recovery was detected for the other two glitches. More interestingly, we found that the leading component of the integral pulse profile after the second glitch became stronger, while the main component became weaker. Our results will expand the sample of pulsars with magnetosphere fluctuation triggered by the glitch event.

astro-ph.HE↗

arXiv2023

CRS-FL: Conditional Random Sampling for Communication-Efficient and Privacy-Preserving Federated Learning

Federated Learning (FL), a privacy-oriented distributed ML paradigm, is being gaining great interest in Internet of Things because of its capability to protect participants data privacy. Studies have been conducted to address challenges existing in standard FL, including communication efficiency and privacy-preserving. But they cannot achieve the goal of making a tradeoff between communication efficiency and model accuracy while guaranteeing privacy. This paper proposes a Conditional Random Sampling (CRS) method and implements it into the standard FL settings (CRS-FL) to tackle the above-mentioned challenges. CRS explores a stochastic coefficient based on Poisson sampling to achieve a higher probability of obtaining zero-gradient unbiasedly, and then decreases the communication overhead effectively without model accuracy degradation. Moreover, we dig out the relaxation Local Differential Privacy (LDP) guarantee conditions of CRS theoretically. Extensive experiment results indicate that (1) in communication efficiency, CRS-FL performs better than the existing methods in metric accuracy per transmission byte without model accuracy reduction in more than 7% sampling ratio (# sampling size / # model size); (2) in privacy-preserving, CRS-FL achieves no accuracy reduction compared with LDP baselines while holding the efficiency, even exceeding them in model accuracy under more sampling ratio conditions.

cs.CR↗

arXiv2023

Robust Training for Speaker Verification against Noisy Labels

The deep learning models used for speaker verification rely heavily on large amounts of data and correct labeling. However, noisy (incorrect) labels often occur, which degrades the performance of the system. In this paper, we propose a novel two-stage learning method to filter out noisy labels from speaker datasets. Since a DNN will first fit data with clean labels, we first train the model with all data for several epochs. Then, based on this model, the model predictions are compared with the labels using our proposed the OR-Gate with top-k mechanism to select the data with clean labels and the selected data is used to train the model. This process is iterated until the training is completed. We have demonstrated the effectiveness of this method in filtering noisy labels through extensive experiments and have achieved excellent performance on the VoxCeleb (1 and 2) with different added noise rates.

cs.SD↗

arXiv2023

Conductance Quantization in PbTe Nanowires

PbTe nanowires coupled to a superconductor have recently been proposed as a potential Majorana platform. The hallmark of the one-dimensional nature of ballistic nanowires is their quantized conductance. Here, we report the observation of conductance plateaus at multiples of the quantized value $2e^2/h$ in PbTe nanowires at finite magnetic fields. The quantized plateaus, as a function of source-drain bias and magnetic field, allow for the extraction of the Landé $g$-factor, sub-band spacing and effective mass. The coefficient of 2 in the plateau conductance indicates the presence of valley degeneracy arising from the crystal orientation of the nanowires, which are grown on a (001) substrate. Occasionally, this degeneracy can be lifted by a gate voltage that breaks the mirror symmetry. Our results demonstrate the one-dimensionality of PbTe nanowires and fulfill one of the necessary conditions for the realization of Majorana zero modes.

cond-mat.mes-hall↗

arXiv2023

Quantized anomalous Hall resistivity achieved in molecular beam epitaxy-grown MnBi2Te4 thin films

The intrinsic magnetic topological insulator MnBi2Te4 provides a feasible pathway to high temperature quantum anomalous Hall (QAH) effect as well as various novel topological quantum phases. Although quantized transport properties have been observed in exfoliated MnBi2Te4 thin flakes, it remains a big challenge to achieve molecular beam epitaxy (MBE)-grown MnBi2Te4 thin films even close to the quantized regime. In this work, we report the realization of quantized anomalous Hall resistivity in MBE-grown MnBi2Te4 thin films with the chemical potential tuned by both controlled in-situ oxygen exposure and top gating. We find that elongated post-annealing obviously elevates the temperature to achieve quantization of the Hall resistivity, but also increases the residual longitudinal resistivity, indicating a picture of high-quality QAH puddles weakly coupled by tunnel barriers. These results help to clarify the puzzles in previous experimental studies on MnBi2Te4 and to find a way out of the big difficulty in obtaining MnBi2Te4 samples showing quantized transport properties.

cond-mat.mes-hall↗

arXiv2023

Improved Adversarial Training Through Adaptive Instance-wise Loss Smoothing

Deep neural networks can be easily fooled into making incorrect predictions through corruption of the input by adversarial perturbations: human-imperceptible artificial noise. So far adversarial training has been the most successful defense against such adversarial attacks. This work focuses on improving adversarial training to boost adversarial robustness. We first analyze, from an instance-wise perspective, how adversarial vulnerability evolves during adversarial training. We find that during training an overall reduction of adversarial loss is achieved by sacrificing a considerable proportion of training samples to be more vulnerable to adversarial attack, which results in an uneven distribution of adversarial vulnerability among data. Such "uneven vulnerability", is prevalent across several popular robust training methods and, more importantly, relates to overfitting in adversarial training. Motivated by this observation, we propose a new adversarial training method: Instance-adaptive Smoothness Enhanced Adversarial Training (ISEAT). It jointly smooths both input and weight loss landscapes in an adaptive, instance-specific, way to enhance robustness more for those samples with higher adversarial vulnerability. Extensive experiments demonstrate the superiority of our method over existing defense methods. Noticeably, our method, when combined with the latest data augmentation and semi-supervised learning techniques, achieves state-of-the-art robustness against $\ell_{\infty}$-norm constrained attacks on CIFAR10 of 59.32% for Wide ResNet34-10 without extra data, and 61.55% for Wide ResNet28-10 with extra data. Code is available at https://github.com/TreeLLi/Instance-adaptive-Smoothness-Enhanced-AT.

cs.CV↗

arXiv2023

The Human-Centric Metaverse: A Survey

In the era of the Web of Things, the Metaverse is expected to be the landing site for the next generation of the Internet, resulting in the increased popularity of related technologies and applications in recent years and gradually becoming the focus of Internet research. The Metaverse, as a link between the real and virtual worlds, can provide users with immersive experiences. As the concept of the Metaverse grows in popularity, many scholars and developers begin to focus on the Metaverse's ethics and core. This paper argues that the Metaverse should be centered on humans. That is, humans constitute the majority of the Metaverse. As a result, we begin this paper by introducing the Metaverse's origins, characteristics, related technologies, and the concept of the human-centric Metaverse (HCM). Second, we discuss the manifestation of human-centric in the Metaverse. Finally, we discuss some current issues in the construction of HCM. In this paper, we provide a detailed review of the applications of human-centric technologies in the Metaverse, as well as the relevant HCM application scenarios. We hope that this paper can provide researchers and developers with some directions and ideas for human-centric Metaverse construction.

cs.HC↗