Query Attack Research Articles

Adversarial Attack and Defense in Deep Ranking.

Deep Neural Network classifiers are vulnerable to adversarial attacks, where an imperceptible perturbation could result in misclassification. However, the vulnerability of DNN-based image ranking systems remains under-explored. In this paper, we propose two attacks against deep ranking systems, i.e., Candidate Attack and Query Attack, that can raise or lower the rank of chosen candidates by adversarial perturbations. Specifically, the expected ranking order is first represented as a set of inequalities. Then a triplet-like objective function is designed to obtain the optimal perturbation. Conversely, an anti-collapse triplet defense is proposed to improve the ranking model robustness against all proposed attacks, where the model learns to prevent the adversarial attack from pulling the positive and negative samples close to each other. To comprehensively measure the empirical adversarial robustness of a ranking model with our defense, we propose an empirical robustness score, which involves a set of representative attacks against ranking models. Our adversarial ranking attacks and defenses are evaluated on MNIST, Fashion-MNIST, CUB200-2011, CARS196, and Stanford Online Products datasets. Experimental results demonstrate that our attacks can effectively compromise a typical deep ranking system. Nevertheless, our defense can significantly improve the ranking system's robustness and simultaneously mitigate a wide range of attacks.

Query Attack by Multi-Identity Surrogates

Deep Neural Networks (DNNs) are acknowledged as vulnerable to adversarial attacks, while the existing black-box attacks require extensive queries on the victim DNN to achieve high success rates. For query-efficiency, surrogate models of the victim are used to generate transferable Adversarial Examples (AEs) because of their <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Gradient Similarity (GS)</i> , i.e., surrogates' attack gradients are similar to the victim's ones. However, it is generally neglected to exploit their similarity on outputs, namely the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Prediction Similarity (PS)</i> , to filter out inefficient queries by surrogates without querying the victim. To jointly utilize and also optimize surrogates' GS and PS, we develop QueryNet, a unified attack framework that can significantly reduce queries. QueryNet creatively attacks by multi-identity surrogates, i.e., crafts several AEs for one sample by different surrogates, and also uses surrogates to decide on the most promising AE for the query. After that, the victim's query feedback is accumulated to optimize not only surrogates' parameters but also their architectures, enhancing both the GS and the PS. Although QueryNet has no access to pre-trained surrogates' prior, it reduces queries by averagely about an order of magnitude compared to alternatives within an acceptable time, according to our comprehensive experiments: 11 victims (including two commercial models) on MNIST/CIFAR10/ImageNet, allowing only 8-bit image queries, and no access to the victim's training data. The code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Sizhe-Chen/QueryNet</uri> .

Towards Robust Neural Rankers with Large Language Model: A Contrastive Training Approach

Pre-trained language model-based neural rankers have been widely applied in information retrieval (IR). However, the robustness issue of current IR models has not received sufficient attention, which could significantly impact the user experience in practical applications. In this study, we focus on the defensive ability of IR models against query attacks while guaranteeing their retrieval performance. We discover that improving the robustness of IR models not only requires a focus on model architecture and training methods but is also closely related to the quality of data. Different from previous research, we use large language models (LLMs) to generate query variations with the same intent, which exhibit richer and more realistic expressions while maintaining consistent query intent. Based on LLM-generated query variations, we propose a novel contrastive training framework that substantially enhances the robustness of IR models to query perturbations. Specifically, we combine the contrastive loss in the representation space of query variations with the ranking loss in the retrieval training stage to improve the model’s ability to understand the underlying semantic information of queries. Experimental results on two public datasets, WikiQA and ANTIQUE, demonstrate that the proposed contrastive training approach effectively improves the robustness of models facing query attack scenarios while outperforming baselines in retrieval performance. Compared with the best baseline approach, the improvements in average robustness performance of Reranker IR models are 24.9%, 26.5%, 27.0%, and 75.0% on WikiQA and 8.7%, 1.9%, 6.3%, and 13.6% on ANTIQUE, in terms of the MAP (Mean Average Precision), MRR (Mean Reciprocal Rank), nDCG@10 (Normalized Discounted Cumulative Gain) and P@10 (Precision), respectively.

Hybrid Protection of Digital FIR Filters

A digital finite impulse response (FIR) filter is a ubiquitous block in digital signal processing applications and its behavior is determined by its coefficients. To protect filter coefficients from an adversary, efficient obfuscation techniques have been proposed, either by hiding them behind decoys or replacing them by key bits. In this article, we initially introduce a query attack that can discover the secret key of such obfuscated FIR filters, which could not be broken by the existing prominent attacks. Then, we propose a first of its kind hybrid technique, including both hardware obfuscation and logic locking using a point function for the protection of parallel direct and transposed forms of digital FIR filters. Experimental results show that the hybrid protection technique can lead to FIR filters with higher security while maintaining the hardware complexity competitive or superior to those locked by prominent logic locking methods. It is also shown that the protected multiplier blocks and FIR filters are resilient to existing attacks. The results on different forms and realizations of FIR filters show that the parallel direct form FIR filter has a promising potential for a secure design.

An Evolutionary, Gradient-Free, Query-Efficient, Black-Box Algorithm for Generating Adversarial Instances in Deep Convolutional Neural Networks

Deep neural networks (DNNs) are sensitive to adversarial data in a variety of scenarios, including the black-box scenario, where the attacker is only allowed to query the trained model and receive an output. Existing black-box methods for creating adversarial instances are costly, often using gradient estimation or training a replacement network. This paper introduces Query-Efficient Evolutionary Attack—QuEry Attack—an untargeted, score-based, black-box attack. QuEry Attack is based on a novel objective function that can be used in gradient-free optimization problems. The attack only requires access to the output logits of the classifier and is thus not affected by gradient masking. No additional information is needed, rendering our method more suitable to real-life situations. We test its performance with three different, commonly used, pretrained image-classifications models—Inception-v3, ResNet-50, and VGG-16-BN—against three benchmark datasets: MNIST, CIFAR10 and ImageNet. Furthermore, we evaluate QuEry Attack’s performance on non-differential transformation defenses and robust models. Our results demonstrate the superior performance of QuEry Attack, both in terms of accuracy score and query efficiency.

Detecting The Normal And Injected Sql Query Using Random Forest Classification With Rabin Carp Pattern Matching Algorithm In Web Database

Nowadays detection of SQL injection attack is becoming very important in database-driven websites. SQL-Injection attack is a class of attacks that many of these systems are highly vulnerable to, and there is no known fool-proof defense against various attacks. In order to overcome this problem, the assurance for security to all the web Sites is essential for the protections for their database. Various Companies facing challenges in intrusion and Detection while installation and deploying their sites. And most common attack in the way if requesting the input as malicious logic in sql query of the authentication page. So that query returns success response in the condition of functionality. In this paper, we proposed the two algorithms such as Random forest and Rabin carp pattern matching to detect the injected query attack. Random forest is the technique used to detect the anomaly queries from the dataset. And Rabin carp pattern matching is used to prevent SQL injection in authentication page.

Retracted] Security Control Technology and Simulation of Network News Communication under the Environment of Internet of Things

Internet of Things is an application of network news communication technology. Based on the Internet, it uses physical access technologies such as radio frequency tag and wireless sensor network news communication and network news communication information transmission technology to build a network news communication information system that can cover people and things. In the physical layer, the relative position of the object is calculated by using multipoint cooperative localization, so as to determine the minimum anonymous region. Generate and maintain the anonymous tree topology on the network news dissemination layer, and provide storage management support for multiple anonymous groups. In the application layer, the object determines the corresponding anonymity degree according to the identity and uses the frame structure to construct and return the new anonymous group consistent with the existing anonymous group, which can prevent the persistent multiprecision query attack. A real‐time control method for intrusion response of a security control system is designed, which includes two stages: response task generation and integrated scheduling. An intrusion response task set generation method based on an improved nondominated sorting genetic algorithm is presented, and a distributed integrated task scheduling and optimization algorithm based on a genetic algorithm and a directed acyclic graph is designed. The numerical simulation results show that this method can quickly and smoothly implement the response strategy of information security intrusion without affecting the normal execution of system tasks.

REPLICATION OF COMPARATIVE METHODS FOR SINGLE DATABASE PERFORMANCE IN CENTOS OPERATION SYSTEM WITH MAXSCALE

Processing data in an institution or organization has become an activity or routine for each meeting. In processing data, of course there are various kinds of problems, ranging from missing data, either because of an error in entering data or the result of a system failure. To overcome these problems, in an agency or company must have data backup, so as to create problems that occur in the data, the institution still has data backup. To call, another method of database replication is used. From the test results obtained data on the value of RTA (Round Trip Average) that is the time required for one time sending and receiving data from the replication database is 5426.22 ms and for compilation service diagrams a query attack occurs with a percentage of 29% in the database applications that produce 1774.45 ms results and for service diagrams at a query attack percentage of 29% a single database estimates the server down .

Novel trajectory privacy-preserving method based on clustering using differential privacy

With the development of location-aware technology, a large amount of location data of users is collected by the trajectory database. If these trajectory data are directly used for data mining without being processed, it will pose a threat to the user's personal privacy. At the moment, differential privacy is favored by experts and scholars because of its strict mathematical rigor, but how to apply differential privacy technology to trajectory clustering analysis is a difficult problem. To solve the problems in which existing trajectory privacy-preserving models have poor data availability or difficulty to resist complex privacy attacks, we devise novel trajectory privacy-preserving method based on clustering using differential privacy. More specifically, Laplacian noise is added to the count of trajectory location in the cluster to resist the continuous query attack. Then, radius-constrained Laplacian noise is added to the trajectory location data in the cluster to avoid too much noise affecting the clustering effect. According to the noise location data and the count of noise location, the noise clustering center in the cluster is obtained. Finally, it is considered that the attacker can associate the user trajectory with other information to form secret reasoning attack, and secret reasoning attack model is proposed. And we use the differential privacy technology to give corresponding resistance. Experimental results using the open data show that the proposed algorithm can not only effectively protect the private information of the trajectory data, but also ensure the data availability in cluster analysis. And compared with other algorithms, our algorithm has good effect on some evaluation indicators.

Achieving High Privacy Protection in Location-based Recommendation Systems

In recent years, privacy has become great attention in the research community. In Location-based Recommendation Systems (LbRSs), the user is constrained to build queries depend on his actual position to search for the closest points of interest (POIs). An external attacker can analyze the sent queries or track the actual position of the LbRS user to reveal his\her personal information. Consequently, ensuring high privacy protection (which is including location privacy and query privacy) is a fundamental thing. In this paper, we propose a model that guarantees high privacy protection for LbRS users. The model is work by three components: The first component (selector) uses a new location privacy protection approach, namely, the smart dummy selection (SDS) approach. The SDS approach generates a strong dummy position that has high resistance versus a semantic position attack. The second component (encryptor) uses an encryption-based approach that guarantees a high level of query privacy versus a sampling query attack. The last component (constructor) constructs the protected query that is sent to the LbRS server. Our proposed model is supported by a checkpoint technique to ensure a high availability quality attribute. Our proposed model yields competitive results compared to similar models under various privacy and performance metrics.

Searchable Encryption over Feature-Rich Data

Storage services allow data owners to store their huge amount of potentially sensitive data, such as audios, images, and videos, on remote cloud servers in encrypted form. To enable retrieval of encrypted files of interest, searchable symmetric encryption (SSE) schemes have been proposed. However, many schemes construct indexes based on keyword-file pairs and focus on boolean expressions of exact keyword matches. Moreover, most dynamic SSE schemes cannot achieve forward privacy and reveal unnecessary information when updating the encrypted databases. We tackle the challenge of supporting large-scale similarity search over encrypted feature-rich multimedia data, by considering the search criteria as a high-dimensional feature vector instead of a keyword. Our solutions are built on carefully-designed fuzzy Bloom filters which utilize locality sensitive hashing (LSH) to encode an index associating the file identifiers and feature vectors. Our schemes are proven to be secure against adaptively chosen query attack and forward private in the standard model . We have evaluated the performance of our scheme on real-world high-dimensional datasets, and achieved a search quality of 99 percent recall with only a few number of hash tables for LSH. This shows that our index is compact and searching is not only efficient but also accurate.

Merkle’s Key Agreement Protocol is Optimal: An $$O(n^2)$$ O ( n 2 ) Attack on Any Key Agreement from Random Oracles

We prove that every key agreement protocol in the random oracle model in which the honest users make at most n queries to the oracle can be broken by an adversary who makes \(O(n^2)\) queries to the oracle. This improves on the previous \({\tilde{\Omega }}(n^6)\) query attack given by Impagliazzo and Rudich (STOC ’89) and resolves an open question posed by them. Our bound is optimal up to a constant factor since Merkle proposed a key agreement protocol in 1974 that can be easily implemented with n queries to a random oracle and cannot be broken by any adversary who asks \(o(n^2)\) queries.

Leakage‐resilient message authentication code scheme based on hidden identity weak hash proof system

Hazay et al. initiated the formal study of leakage-resilient message authentication code (MAC) and presented a MAC scheme that is both leakage resilient and unforgeable against chosen message and no verification query attack (uf-cm-nvq). As the communication overhead of their construction is linear with the parameters that control the leakage bound, their scheme sacrifices efficiency in exchange for leakage resilience. In this study, the authors study the problem of designing leakage-resilient MACs in the public-key setting with acceptable communication efficiency. In particular, a notion called ‘hidden identity weak hash proof system’(HID-wHPS) is introduced. Then a generic MAC construction is presented under the abstraction framework of HID-wHPS. Security properties guaranteed by HID-wHPS enable us to prove the author's construction to be both leakage resilient and uf-cm-nvq in a modular way. Finally, performance analysis shows that their MAC construction yields improved tagging-key size, tag size as well as computation overhead under the given leakage bound.

New scheme of ID-based authenticated multi-party key agreement

Authenticated key agreement protocol allows a group of users in an open network environment to identify each other and share a security session key.This article proposed a new scheme of ID-based authenticated multi-party key agreement based on McCullagh-Barreto scheme.Key seed was introduced to update temporary public/private key pairs.The new scheme is able to realize the authentication,improve the security,resist Reveal query attack and the key compromise impersonation attack successfully,and it has many properties such as non-key control and equal contribution.