arXiv subjects

Xiaojun Jia

Publications and source records attributed to Xiaojun Jia.

At least 127 records · Page 7Linked to original sources

arXiv2023

Improving Fast Adversarial Training with Prior-Guided Knowledge

Fast adversarial training (FAT) is an efficient method to improve robustness. However, the original FAT suffers from catastrophic overfitting, which dramatically and suddenly reduces robustness after a few training epochs. Although various FAT variants have been proposed to prevent overfitting, they require high training costs. In this paper, we investigate the relationship between adversarial example quality and catastrophic overfitting by comparing the training processes of standard adversarial training and FAT. We find that catastrophic overfitting occurs when the attack success rate of adversarial examples becomes worse. Based on this observation, we propose a positive prior-guided adversarial initialization to prevent overfitting by improving adversarial example quality without extra training costs. This initialization is generated by using high-quality adversarial perturbations from the historical training process. We provide theoretical analysis for the proposed initialization and propose a prior-guided regularization method that boosts the smoothness of the loss function. Additionally, we design a prior-guided ensemble FAT method that averages the different model weights of historical models using different decay rates. Our proposed method, called FGSM-PGK, assembles the prior-guided knowledge, i.e., the prior-guided initialization and model weights, acquired during the historical training process. Evaluations of four datasets demonstrate the superiority of the proposed method.

cs.LG↗

arXiv2022

Context-Aware Robust Fine-Tuning

Contrastive Language-Image Pre-trained (CLIP) models have zero-shot ability of classifying an image belonging to "[CLASS]" by using similarity between the image and the prompt sentence "a [CONTEXT] of [CLASS]". Based on exhaustive text cues in "[CONTEXT]", CLIP model is aware of different contexts, e.g. background, style, viewpoint, and exhibits unprecedented robustness against a wide range of distribution shifts. However, recent works find further fine-tuning of CLIP models improves accuracy but sacrifices the robustness on downstream tasks. We conduct an empirical investigation to show fine-tuning will corrupt the context-aware ability of pre-trained CLIP features. To solve this problem, we propose Context-Aware Robust Fine-tuning (CAR-FT). CAR-FT regularizes the model during fine-tuning to capture the context information. Specifically, we use zero-shot prompt weights to get the context distribution contained in the image. By minimizing the Kullback-Leibler Divergence (KLD) between context distributions induced by original/fine-tuned CLIP models, CAR-FT makes the context-aware ability of CLIP inherited into downstream tasks, and achieves both higher In-Distribution (ID) and Out-Of-Distribution (OOD) accuracy. The experimental results show CAR-FT achieves superior robustness on five OOD test datasets of ImageNet, and meanwhile brings accuracy gains on nine downstream tasks. Additionally, CAR-FT surpasses previous Domain Generalization (DG) methods and gets 78.5% averaged accuracy on DomainBed benchmark, building the new state-of-the-art.

cs.CV↗

arXiv2022

A Large-scale Multiple-objective Method for Black-box Attack against Object Detection

Recent studies have shown that detectors based on deep models are vulnerable to adversarial examples, even in the black-box scenario where the attacker cannot access the model information. Most existing attack methods aim to minimize the true positive rate, which often shows poor attack performance, as another sub-optimal bounding box may be detected around the attacked bounding box to be the new true positive one. To settle this challenge, we propose to minimize the true positive rate and maximize the false positive rate, which can encourage more false positive objects to block the generation of new true positive bounding boxes. It is modeled as a multi-objective optimization (MOP) problem, of which the generic algorithm can search the Pareto-optimal. However, our task has more than two million decision variables, leading to low searching efficiency. Thus, we extend the standard Genetic Algorithm with Random Subset selection and Divide-and-Conquer, called GARSDC, which significantly improves the efficiency. Moreover, to alleviate the sensitivity to population quality in generic algorithms, we generate a gradient-prior initial population, utilizing the transferability between different detectors with similar backbones. Compared with the state-of-art attack methods, GARSDC decreases by an average 12.0 in the mAP and queries by about 1000 times in extensive experiments. Our codes can be found at https://github.com/LiangSiyuan21/ GARSDC.

cs.CV↗

arXiv2022

Prior-Guided Adversarial Initialization for Fast Adversarial Training

Fast adversarial training (FAT) effectively improves the efficiency of standard adversarial training (SAT). However, initial FAT encounters catastrophic overfitting, i.e.,the robust accuracy against adversarial attacks suddenly and dramatically decreases. Though several FAT variants spare no effort to prevent overfitting, they sacrifice much calculation cost. In this paper, we explore the difference between the training processes of SAT and FAT and observe that the attack success rate of adversarial examples (AEs) of FAT gets worse gradually in the late training stage, resulting in overfitting. The AEs are generated by the fast gradient sign method (FGSM) with a zero or random initialization. Based on the observation, we propose a prior-guided FGSM initialization method to avoid overfitting after investigating several initialization strategies, improving the quality of the AEs during the whole training process. The initialization is formed by leveraging historically generated AEs without additional calculation cost. We further provide a theoretical analysis for the proposed initialization method. We also propose a simple yet effective regularizer based on the prior-guided initialization,i.e., the currently generated perturbation should not deviate too much from the prior-guided initialization. The regularizer adopts both historical and current adversarial perturbations to guide the model learning. Evaluations on four datasets demonstrate that the proposed method can prevent catastrophic overfitting and outperform state-of-the-art FAT methods. The code is released at https://github.com/jiaxiaojunQAQ/FGSM-PGI.

cs.CV↗

arXiv2022

Watermark Vaccine: Adversarial Attacks to Prevent Watermark Removal

As a common security tool, visible watermarking has been widely applied to protect copyrights of digital images. However, recent works have shown that visible watermarks can be removed by DNNs without damaging their host images. Such watermark-removal techniques pose a great threat to the ownership of images. Inspired by the vulnerability of DNNs on adversarial perturbations, we propose a novel defence mechanism by adversarial machine learning for good. From the perspective of the adversary, blind watermark-removal networks can be posed as our target models; then we actually optimize an imperceptible adversarial perturbation on the host images to proactively attack against watermark-removal networks, dubbed Watermark Vaccine. Specifically, two types of vaccines are proposed. Disrupting Watermark Vaccine (DWV) induces to ruin the host image along with watermark after passing through watermark-removal networks. In contrast, Inerasable Watermark Vaccine (IWV) works in another fashion of trying to keep the watermark not removed and still noticeable. Extensive experiments demonstrate the effectiveness of our DWV/IWV in preventing watermark removal, especially on various watermark removal networks.

cs.CV↗

arXiv2022

High-performance cavity-enhanced quantum memory with warm atomic cell

High-performance quantum memory for quantized states of light is a prerequisite building block of quantum information technology. Despite great progresses of optical quantum memories based on interactions of light and atoms, physical features of these memories still cannot satisfy requirements for applications in practical quantum information systems, since all of them suffer from trade-off between memory efficiency and excess noise. Here, we report a high-performance cavity-enhanced electromagnetically-induced-transparency memory with warm atomic cell in which a scheme of optimizing the spatial and temporal modes based on the time-reversal approach is applied. The memory efficiency up to 67% is directly measured and a noise level close to quantum noise limit is simultaneously reached. It has been experimentally demonstrated that the average fidelities for a set of input coherent states with different phases and amplitudes within a Gaussian distribution have exceeded the classical benchmark fidelities. Thus the realized quantum memory platform has been capable of preserving quantized optical states, and is ready to be applied in quantum information systems, such as distributed quantum logic gates and quantum-enhanced atomic magnetometry.

quant-ph↗

arXiv2022

Boosting Fast Adversarial Training with Learnable Adversarial Initialization

Adversarial training (AT) has been demonstrated to be effective in improving model robustness by leveraging adversarial examples for training. However, most AT methods are in face of expensive time and computational cost for calculating gradients at multiple steps in generating adversarial examples. To boost training efficiency, fast gradient sign method (FGSM) is adopted in fast AT methods by calculating gradient only once. Unfortunately, the robustness is far from satisfactory. One reason may arise from the initialization fashion. Existing fast AT generally uses a random sample-agnostic initialization, which facilitates the efficiency yet hinders a further robustness improvement. Up to now, the initialization in fast AT is still not extensively explored. In this paper, we boost fast AT with a sample-dependent adversarial initialization, i.e., an output from a generative network conditioned on a benign image and its gradient information from the target network. As the generative network and the target network are optimized jointly in the training phase, the former can adaptively generate an effective initialization with respect to the latter, which motivates gradually improved robustness. Experimental evaluations on four benchmark databases demonstrate the superiority of our proposed method over state-of-the-art fast AT methods, as well as comparable robustness to advanced multi-step AT methods. The code is released at https://github.com//jiaxiaojunQAQ//FGSM-SDI.

cs.CV↗

arXiv2022

LAS-AT: Adversarial Training with Learnable Attack Strategy

Adversarial training (AT) is always formulated as a minimax problem, of which the performance depends on the inner optimization that involves the generation of adversarial examples (AEs). Most previous methods adopt Projected Gradient Decent (PGD) with manually specifying attack parameters for AE generation. A combination of the attack parameters can be referred to as an attack strategy. Several works have revealed that using a fixed attack strategy to generate AEs during the whole training phase limits the model robustness and propose to exploit different attack strategies at different training stages to improve robustness. But those multi-stage hand-crafted attack strategies need much domain expertise, and the robustness improvement is limited. In this paper, we propose a novel framework for adversarial training by introducing the concept of "learnable attack strategy", dubbed LAS-AT, which learns to automatically produce attack strategies to improve the model robustness. Our framework is composed of a target network that uses AEs for training to improve robustness and a strategy network that produces attack strategies to control the AE generation. Experimental evaluations on three benchmark databases demonstrate the superiority of the proposed method. The code is released at https://github.com/jiaxiaojunQAQ/LAS-AT.

cs.CV↗

arXiv2021

Defending against Model Stealing via Verifying Embedded External Features

Obtaining a well-trained model involves expensive data collection and training procedures, therefore the model is a valuable intellectual property. Recent studies revealed that adversaries can `steal' deployed models even when they have no training samples and can not get access to the model parameters or structures. Currently, there were some defense methods to alleviate this threat, mostly by increasing the cost of model stealing. In this paper, we explore the defense from another angle by verifying whether a suspicious model contains the knowledge of defender-specified \emph{external features}. Specifically, we embed the external features by tempering a few training samples with style transfer. We then train a meta-classifier to determine whether a model is stolen from the victim. This approach is inspired by the understanding that the stolen models should contain the knowledge of features learned by the victim model. We examine our method on both CIFAR-10 and ImageNet datasets. Experimental results demonstrate that our method is effective in detecting different types of model stealing simultaneously, even if the stolen model is obtained via a multi-stage stealing process. The codes for reproducing main results are available at Github (https://github.com/zlh-thu/StealingVerification).

cs.CR↗

arXiv2021

An Effective and Robust Detector for Logo Detection

In recent years, intellectual property (IP), which represents literary, inventions, artistic works, etc, gradually attract more and more people's attention. Particularly, with the rise of e-commerce, the IP not only represents the product design and brands, but also represents the images/videos displayed on e-commerce platforms. Unfortunately, some attackers adopt some adversarial methods to fool the well-trained logo detection model for infringement. To overcome this problem, a novel logo detector based on the mechanism of looking and thinking twice is proposed in this paper for robust logo detection. The proposed detector is different from other mainstream detectors, which can effectively detect small objects, long-tail objects, and is robust to adversarial images. In detail, we extend detectoRS algorithm to a cascade schema with an equalization loss function, multi-scale transformations, and adversarial data augmentation. A series of experimental results have shown that the proposed method can effectively improve the robustness of the detection model. Moreover, we have applied the proposed methods to competition ACM MM2021 Robust Logo Detection that is organized by Alibaba on the Tianchi platform and won top 2 in 36489 teams. Code is available at https://github.com/jiaxiaojunQAQ/Robust-Logo-Detection.

cs.CV↗

arXiv2020

Adv-watermark: A Novel Watermark Perturbation for Adversarial Examples

Recent research has demonstrated that adding some imperceptible perturbations to original images can fool deep learning models. However, the current adversarial perturbations are usually shown in the form of noises, and thus have no practical meaning. Image watermark is a technique widely used for copyright protection. We can regard image watermark as a king of meaningful noises and adding it to the original image will not affect people's understanding of the image content, and will not arouse people's suspicion. Therefore, it will be interesting to generate adversarial examples using watermarks. In this paper, we propose a novel watermark perturbation for adversarial examples (Adv-watermark) which combines image watermarking techniques and adversarial example algorithms. Adding a meaningful watermark to the clean images can attack the DNN models. Specifically, we propose a novel optimization algorithm, which is called Basin Hopping Evolution (BHE), to generate adversarial watermarks in the black-box attack mode. Thanks to the BHE, Adv-watermark only requires a few queries from the threat models to finish the attacks. A series of experiments conducted on ImageNet and CASIA-WebFace datasets show that the proposed method can efficiently generate adversarial examples, and outperforms the state-of-the-art attack methods. Moreover, Adv-watermark is more robust against image transformation defense methods.

cs.CR↗

arXiv2020

Quantum interferometer combining squeezing and parametric amplification

High precision interferometers are the building blocks of precision metrology and the ultimate interferometric sensitivity is limited by the quantum noise. Here we propose and experimentally demonstrate a compact quantum interferometer involving two optical parametric amplifiers and the squeezed states generated within the interferometer are directly used for the phase-sensing quantum state. By both squeezing shot noise and amplifying phase-sensing intensity the sensitivity improvement of $4.86\pm 0.24$ dB beyond the standard quantum limit is deterministically realized and a minimum detectable phase smaller than that of all present interferometers under the same phase-sensing intensity is achieved. This interferometric system has significantly potential applications in a variety of measurements for tiny variances of physical quantities.

quant-ph↗

arXiv2020

Quantum enhanced optical phase estimation with a squeezed thermal state

Quantum phase estimation protocols can provide a measuring method of phase shift with precision superior to standard quantum limit (SQL) due to the application of a nonclassical state of light. A squeezed vacuum state, whose variance in one quadrature is lower than the corresponding SQL, has been pointed out a sensitive resource for quantum phase estimation and the estimation accuracy is directly influenced by the properties of the squeezed state. Here we detailedly analyze the influence of the purity and squeezing level of the squeezed state on the accuracy of quantum phase estimation. The maximum precision that can be achieved for a squeezed thermal state is evaluated, and the experimental results are in agreement with the theoretical analyses. It is also found that the width of the phase estimation interval $Δθ$ beyond SQL is correlated with the purity of the squeezed state.

quant-ph↗

arXiv2019

Identifying and Resisting Adversarial Videos Using Temporal Consistency

Video classification is a challenging task in computer vision. Although Deep Neural Networks (DNNs) have achieved excellent performance in video classification, recent research shows adding imperceptible perturbations to clean videos can make the well-trained models output wrong labels with high confidence. In this paper, we propose an effective defense framework to characterize and defend adversarial videos. The proposed method contains two phases: (1) adversarial video detection using temporal consistency between adjacent frames, and (2) adversarial perturbation reduction via denoisers in the spatial and temporal domains respectively. Specifically, because of the linear nature of DNNs, the imperceptible perturbations will enlarge with the increasing of DNNs depth, which leads to the inconsistency of DNNs output between adjacent frames. However, the benign video frames often have the same outputs with their neighbor frames owing to the slight changes. Based on this observation, we can distinguish between adversarial videos and benign videos. After that, we utilize different defense strategies against different attacks. We propose the temporal defense, which reconstructs the polluted frames with their temporally neighbor clean frames, to deal with the adversarial videos with sparse polluted frames. For the videos with dense polluted frames, we use an efficient adversarial denoiser to process each frame in the spatial domain, and thus purify the perturbations (we call it as spatial defense). A series of experiments conducted on the UCF-101 dataset demonstrate that the proposed method significantly improves the robustness of video classifiers against adversarial attacks.

cs.LG↗

arXiv2019

ComDefend: An Efficient Image Compression Model to Defend Adversarial Examples

Deep neural networks (DNNs) have been demonstrated to be vulnerable to adversarial examples. Specifically, adding imperceptible perturbations to clean images can fool the well trained deep neural networks. In this paper, we propose an end-to-end image compression model to defend adversarial examples: \textbf{ComDefend}. The proposed model consists of a compression convolutional neural network (ComCNN) and a reconstruction convolutional neural network (ResCNN). The ComCNN is used to maintain the structure information of the original image and purify adversarial perturbations. And the ResCNN is used to reconstruct the original image with high quality. In other words, ComDefend can transform the adversarial image to its clean version, which is then fed to the trained classifier. Our method is a pre-processing module, and does not modify the classifier's structure during the whole process. Therefore, it can be combined with other model-specific defense models to jointly improve the classifier's robustness. A series of experiments conducted on MNIST, CIFAR10 and ImageNet show that the proposed method outperforms the state-of-the-art defense methods, and is consistently effective to protect classifiers against adversarial attacks.

cs.CV↗

arXiv2019

Preparation of bipartite bound entangled Gaussian states in quantum optics

The positivity of the partial transpose is in general only a necessary condition for separability. There exist quantum states that are not separable, but nevertheless are positive under partial transpose. States of this type are known as bound entangled states meaning that these states are entangled but they do not allow distillation of pure entanglement by means of local operations and classical communication (LOCC). We present a parametrization of a class of $2\times 2$ bound entangled Gaussian states for bipartite continuous-variable quantum systems with two modes on each side. We propose an experimental protocol for preparing a particular bound entangled state in quantum optics. We then discuss the robustness properties of this protocol with respect to the occupation number of thermal inputs and the degrees of squeezing.

quant-ph↗

arXiv2018

Compact sub-kilohertz low-frequency quantum light source based on four-wave mixing in cesium vapor

Using a nondegenerate four-wave mixing (FWM) process based on a double-Λ scheme in hot cesium vapor, we demonstrate a compact diode-laser-pumped quantum light source for the generation of quantum correlated twin beams with a maximum squeezing of 6.5 dB. The squeezing is observed at a Fourier frequency in the audio band down to 0.7 kHz which, to the best of our knowledge, is the first observation of sub-kilohertz intensity-difference squeezing in an atomic system so far. A phase-matching condition is also investigated in our system, which confirms the spatial-multi-mode characteristics of the FWM process. Our compact low-frequency squeezed light source may find applications in quantum imaging, quantum metrology, and the transfer of optical squeezing onto a matter wave.

quant-ph↗

arXiv2017

Establishing and storing of deterministic quantum entanglement among three distant atomic ensembles

It is crucial for physical realization of quantum information networks to first establish entanglement among multiple space-separated quantum memories and then at a user-controlled moment to transfer the stored entanglement to quantum channels for distribution and conveyance of information. Here we present an experimental demonstration on generation, storage and transfer of deterministic quantum entanglement among three spatially separated atomic ensembles. The off-line prepared multipartite entanglement of optical modes is mapped into three distant atomic ensembles to establish entanglement of atomic spin waves via electromagnetically-induced-transparency light-matter interaction. Then the stored atomic entanglement is transferred into a tripartite quadrature entangled state of light, which is space-separated and can be dynamically allocated to three quantum channels for conveying quantum information. The existence of entanglement among released three optical modes verifies that the system has capacity of preserving multipartite entanglement. The presented protocol can be directly extended to larger quantum networks with more nodes.

quant-ph↗