Brainwaves have demonstrated to be unique enough across individuals to be useful as biometrics. They also provide promising advantages over traditional means of authentication, such as resistance to external observability, revocability, and intrinsic liveness detection. However, most of the research so far has been conducted with expensive, bulky, medical-grade helmets, which offer limited applicability for everyday usage. With the aim to bring brainwave authentication and its benefits closer to real world deployment, we investigate brain biometrics with consumer devices. We conduct a comprehensive measurement experiment and user study that compare five authentication tasks on a user sample up to 10 times larger than those from previous studies, introducing three novel techniques based on cognitive semantic processing. Furthermore, we apply our analysis on high-quality open brainwave data obtained with a medical-grade headset, to assess the differences. We investigate both the performance, security, and usability of the different options and use this evidence to elicit design and research recommendations. Our results show that it is possible to achieve Equal Error Rates as low as 7.2% (a reduction between 68–72% with respect to existing approaches) based on brain responses to images with current inexpensive technology. We show that the common practice of testing authentication systems only with known attacker data is unrealistic and may lead to overly optimistic evaluations. With regard to adoption, users call for simpler devices, faster authentication, and better privacy.
{"title":"Performance and Usability Evaluation of Brainwave Authentication Techniques with Consumer Devices","authors":"Patricia Arias-Cabarcos, Matin Fallahi, Thilo Habrich, Karen Schulze, Christian Becker, Thorsten Strufe","doi":"https://dl.acm.org/doi/10.1145/3579356","DOIUrl":"https://doi.org/https://dl.acm.org/doi/10.1145/3579356","url":null,"abstract":"<p>Brainwaves have demonstrated to be unique enough across individuals to be useful as biometrics. They also provide promising advantages over traditional means of authentication, such as resistance to external observability, revocability, and intrinsic liveness detection. However, most of the research so far has been conducted with expensive, bulky, medical-grade helmets, which offer limited applicability for everyday usage. With the aim to bring brainwave authentication and its benefits closer to real world deployment, we investigate brain biometrics with consumer devices. We conduct a comprehensive measurement experiment and user study that compare five authentication tasks on a user sample up to 10 times larger than those from previous studies, introducing three novel techniques based on cognitive semantic processing. Furthermore, we apply our analysis on high-quality open brainwave data obtained with a medical-grade headset, to assess the differences. We investigate both the performance, security, and usability of the different options and use this evidence to elicit design and research recommendations. Our results show that it is possible to achieve Equal Error Rates as low as 7.2% (a reduction between 68–72% with respect to existing approaches) based on brain responses to images with current inexpensive technology. We show that the common practice of testing authentication systems only with known attacker data is unrealistic and may lead to overly optimistic evaluations. With regard to adoption, users call for simpler devices, faster authentication, and better privacy. </p><p></p>","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"444 1","pages":""},"PeriodicalIF":2.3,"publicationDate":"2023-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138540652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-13DOI: https://dl.acm.org/doi/10.1145/3575796
Seoyeon Hwang, Ercan Ozturk, Gene Tsudik
Exciting recent advances in genome sequencing, coupled with greatly reduced storage and computation costs, make genomic testing increasingly accessible to individuals. Already today, one’s digitized DNA can be easily obtained from a sequencing lab and later used to conduct numerous tests by engaging with a testing facility. Due to the inherent sensitivity of genetic material and the often-proprietary nature of genomic tests, privacy is a natural and crucial issue. While genomic privacy received a great deal of attention within and outside the research community, genomic security has not been sufficiently studied. This is surprising since the usage of fake or altered genomes can have grave consequences, such as erroneous drug prescriptions and genetic test outcomes.
Unfortunately, in the genomic domain, privacy and security (as often happens) are at odds with each other. In this article, we attempt to reconcile security with privacy in genomic testing by designing a novel technique for a secure and private genomic range query protocol between a genomic testing facility and an individual user. The proposed technique ensures authenticity and completeness of user-supplied genomic material while maintaining its privacy by releasing only the minimum thereof. To confirm its broad usability, we show how to apply the proposed technique to a previously proposed genomic private substring matching protocol. Experiments show that the proposed technique offers good performance and is quite practical. Furthermore, we generalize the genomic range query problem to sparse integer sets and discuss potential use cases.
{"title":"Balancing Security and Privacy in Genomic Range Queries","authors":"Seoyeon Hwang, Ercan Ozturk, Gene Tsudik","doi":"https://dl.acm.org/doi/10.1145/3575796","DOIUrl":"https://doi.org/https://dl.acm.org/doi/10.1145/3575796","url":null,"abstract":"<p>Exciting recent advances in genome sequencing, coupled with greatly reduced storage and computation costs, make genomic testing increasingly accessible to individuals. Already today, one’s digitized DNA can be easily obtained from a sequencing lab and later used to conduct numerous tests by engaging with a testing facility. Due to the inherent sensitivity of genetic material and the often-proprietary nature of genomic tests, privacy is a natural and crucial issue. While genomic privacy received a great deal of attention within and outside the research community, genomic security has not been sufficiently studied. This is surprising since the usage of fake or altered genomes can have grave consequences, such as erroneous drug prescriptions and genetic test outcomes.</p><p>Unfortunately, in the genomic domain, privacy and security (as often happens) are at odds with each other. In this article, we attempt to reconcile security with privacy in genomic testing by designing a novel technique for a secure and private genomic range query protocol between a genomic testing facility and an individual user. The proposed technique ensures <i>authenticity</i> and <i>completeness</i> of user-supplied genomic material while maintaining its <i>privacy</i> by releasing only the minimum thereof. To confirm its broad usability, we show how to apply the proposed technique to a previously proposed genomic private substring matching protocol. Experiments show that the proposed technique offers good performance and is quite practical. Furthermore, we generalize the genomic range query problem to sparse integer sets and discuss potential use cases.</p>","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"9 1","pages":""},"PeriodicalIF":2.3,"publicationDate":"2023-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138540710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Litao Li, Steven H. H. Ding, Yuan Tian, B. Fung, P. Charland, Weihan Ou, Leo Song, Congwei Chen
Software vulnerabilities have been posing tremendous reliability threats to the general public as well as critical infrastructures, and there have been many studies aiming to detect and mitigate software defects at the binary level. Most of the standard practices leverage both static and dynamic analysis, which have several drawbacks like heavy manual workload and high complexity. Existing deep learning-based solutions not only suffer to capture the complex relationships among different variables from raw binary code but also lack the explainability required for humans to verify, evaluate, and patch the detected bugs. We propose VulANalyzeR, a deep learning-based model, for automated binary vulnerability detection, Common Weakness Enumeration-type classification, and root cause analysis to enhance safety and security. VulANalyzeR features sequential and topological learning through recurrent units and graph convolution to simulate how a program is executed. The attention mechanism is integrated throughout the model, which shows how different instructions and the corresponding states contribute to the final classification. It also classifies the specific vulnerability type through multi-task learning as this not only provides further explanation but also allows faster patching for zero-day vulnerabilities. We show that VulANalyzeR achieves better performance for vulnerability detection over the state-of-the-art baselines. Additionally, a Common Vulnerability Exposure dataset is used to evaluate real complex vulnerabilities. We conduct case studies to show that VulANalyzeR is able to accurately identify the instructions and basic blocks that cause the vulnerability even without given any prior knowledge related to the locations during the training phase.
{"title":"VulANalyzeR: Explainable Binary Vulnerability Detection with Multi-task Learning and Attentional Graph Convolution","authors":"Litao Li, Steven H. H. Ding, Yuan Tian, B. Fung, P. Charland, Weihan Ou, Leo Song, Congwei Chen","doi":"10.1145/3585386","DOIUrl":"https://doi.org/10.1145/3585386","url":null,"abstract":"Software vulnerabilities have been posing tremendous reliability threats to the general public as well as critical infrastructures, and there have been many studies aiming to detect and mitigate software defects at the binary level. Most of the standard practices leverage both static and dynamic analysis, which have several drawbacks like heavy manual workload and high complexity. Existing deep learning-based solutions not only suffer to capture the complex relationships among different variables from raw binary code but also lack the explainability required for humans to verify, evaluate, and patch the detected bugs. We propose VulANalyzeR, a deep learning-based model, for automated binary vulnerability detection, Common Weakness Enumeration-type classification, and root cause analysis to enhance safety and security. VulANalyzeR features sequential and topological learning through recurrent units and graph convolution to simulate how a program is executed. The attention mechanism is integrated throughout the model, which shows how different instructions and the corresponding states contribute to the final classification. It also classifies the specific vulnerability type through multi-task learning as this not only provides further explanation but also allows faster patching for zero-day vulnerabilities. We show that VulANalyzeR achieves better performance for vulnerability detection over the state-of-the-art baselines. Additionally, a Common Vulnerability Exposure dataset is used to evaluate real complex vulnerabilities. We conduct case studies to show that VulANalyzeR is able to accurately identify the instructions and basic blocks that cause the vulnerability even without given any prior knowledge related to the locations during the training phase.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"26 1","pages":"1 - 25"},"PeriodicalIF":2.3,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49141878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Adopting mobile healthcare network (MHN) services such as disease detection is fraught with concerns about the security and privacy of the entities involved and the resource restrictions at the Internet of Things (IoT) nodes. Hence, the essential requirements for disease detection services are to (i) produce accurate and fast disease detection without jeopardizing the privacy of health clouds and medical users and (ii) reduce the computational and transmission overhead (energy consumption) of the IoT devices while maintaining the privacy. For privacy preservation of widely used neural network– (NN) based disease detection, existing literature suggests either computationally heavy public key fully homomorphic encryption (FHE), or secure multiparty computation, with a large number of interactions. Hence, the existing privacy-preserving NN schemes are energy consuming and not suitable for resource-constrained IoT nodes in MHN. This work proposes a lightweight, fully homomorphic, symmetric key FHE scheme (SkFhe) to address the issues involved in implementing privacy-preserving NN. Based on SkFhe, widely used non-linear activation functions ReLU and Leaky ReLU are implemented over the encrypted domain. Furthermore, based on the proposed privacy-preserving linear transformation and non-linear activation functions, an energy-efficient, accurate, and privacy-preserving NN is proposed. The proposed scheme guarantees privacy preservation of the health cloud’s NN model and medical user’s data. The experimental analysis demonstrates that the proposed solution dramatically reduces the overhead in communication and computation at the user side compared to the existing schemes. Moreover, the improved energy efficiency at the user is accomplished with reduced diagnosis time without sacrificing classification accuracy.
{"title":"Energy Efficient and Secure Neural Network–based Disease Detection Framework for Mobile Healthcare Network","authors":"Sona Alex, Kirubai Dhanaraj, P. P. Deephi","doi":"10.1145/3585536","DOIUrl":"https://doi.org/10.1145/3585536","url":null,"abstract":"Adopting mobile healthcare network (MHN) services such as disease detection is fraught with concerns about the security and privacy of the entities involved and the resource restrictions at the Internet of Things (IoT) nodes. Hence, the essential requirements for disease detection services are to (i) produce accurate and fast disease detection without jeopardizing the privacy of health clouds and medical users and (ii) reduce the computational and transmission overhead (energy consumption) of the IoT devices while maintaining the privacy. For privacy preservation of widely used neural network– (NN) based disease detection, existing literature suggests either computationally heavy public key fully homomorphic encryption (FHE), or secure multiparty computation, with a large number of interactions. Hence, the existing privacy-preserving NN schemes are energy consuming and not suitable for resource-constrained IoT nodes in MHN. This work proposes a lightweight, fully homomorphic, symmetric key FHE scheme (SkFhe) to address the issues involved in implementing privacy-preserving NN. Based on SkFhe, widely used non-linear activation functions ReLU and Leaky ReLU are implemented over the encrypted domain. Furthermore, based on the proposed privacy-preserving linear transformation and non-linear activation functions, an energy-efficient, accurate, and privacy-preserving NN is proposed. The proposed scheme guarantees privacy preservation of the health cloud’s NN model and medical user’s data. The experimental analysis demonstrates that the proposed solution dramatically reduces the overhead in communication and computation at the user side compared to the existing schemes. Moreover, the improved energy efficiency at the user is accomplished with reduced diagnosis time without sacrificing classification accuracy.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"26 1","pages":"1 - 27"},"PeriodicalIF":2.3,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43263605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Communication networks like the Internet form a large distributed system where a huge number of components run in parallel, such as security protocols and distributed web applications. For what concerns security, it is obviously infeasible to verify them all at once as one monolithic entity; rather, one has to verify individual components in isolation. While many typical components like TLS have been studied intensively, there exists much less research on analyzing and ensuring the security of the composition of security protocols. This is a problem since the composition of systems that are secure in isolation can easily be insecure. The main goal of compositionality is thus a theorem of the form: given a set of components that are already proved secure in isolation and that satisfy a number of easy-to-check conditions, then also their parallel composition is secure. Said conditions should of course also be realistic in practice, or better yet, already be satisfied for many existing components. Another benefit of compositionality is that when one would like to exchange a component with another one, all that is needed is the proof that the new component is secure in isolation and satisfies the composition conditions—without having to re-prove anything about the other components. This article has three contributions over previous work in parallel compositionality. First, we extend the compositionality paradigm to stateful systems: while previous approaches work only for simple protocols that only have a local session state, our result supports participants who maintain long-term databases that can be shared among several protocols. This includes a paradigm for declassification of shared secrets. This result is in fact so general that it also covers many forms of sequential composition as a special case of stateful parallel composition. Second, our compositionality result is formalized and proved in Isabelle/HOL, providing a strong correctness guarantee of our proofs. This also means that one can prove, without gaps, the security of an entire system in Isabelle/HOL, namely the security of components in isolation and the composition conditions, and thus derive the security of the entire system as an Isabelle theorem. For the components one can also make use of our tool PSPSP that can perform automatic proofs for many stateful protocols. Third, for the compositionality conditions we have also implemented an automated check procedure in Isabelle.
{"title":"Stateful Protocol Composition in Isabelle/HOL","authors":"Andreas V. Hess, S. Mödersheim, Achim D. Brucker","doi":"10.1145/3577020","DOIUrl":"https://doi.org/10.1145/3577020","url":null,"abstract":"Communication networks like the Internet form a large distributed system where a huge number of components run in parallel, such as security protocols and distributed web applications. For what concerns security, it is obviously infeasible to verify them all at once as one monolithic entity; rather, one has to verify individual components in isolation. While many typical components like TLS have been studied intensively, there exists much less research on analyzing and ensuring the security of the composition of security protocols. This is a problem since the composition of systems that are secure in isolation can easily be insecure. The main goal of compositionality is thus a theorem of the form: given a set of components that are already proved secure in isolation and that satisfy a number of easy-to-check conditions, then also their parallel composition is secure. Said conditions should of course also be realistic in practice, or better yet, already be satisfied for many existing components. Another benefit of compositionality is that when one would like to exchange a component with another one, all that is needed is the proof that the new component is secure in isolation and satisfies the composition conditions—without having to re-prove anything about the other components. This article has three contributions over previous work in parallel compositionality. First, we extend the compositionality paradigm to stateful systems: while previous approaches work only for simple protocols that only have a local session state, our result supports participants who maintain long-term databases that can be shared among several protocols. This includes a paradigm for declassification of shared secrets. This result is in fact so general that it also covers many forms of sequential composition as a special case of stateful parallel composition. Second, our compositionality result is formalized and proved in Isabelle/HOL, providing a strong correctness guarantee of our proofs. This also means that one can prove, without gaps, the security of an entire system in Isabelle/HOL, namely the security of components in isolation and the composition conditions, and thus derive the security of the entire system as an Isabelle theorem. For the components one can also make use of our tool PSPSP that can perform automatic proofs for many stateful protocols. Third, for the compositionality conditions we have also implemented an automated check procedure in Isabelle.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":" ","pages":"1 - 36"},"PeriodicalIF":2.3,"publicationDate":"2023-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42345755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Brainwaves have demonstrated to be unique enough across individuals to be useful as biometrics. They also provide promising advantages over traditional means of authentication, such as resistance to external observability, revocability, and intrinsic liveness detection. However, most of the research so far has been conducted with expensive, bulky, medical-grade helmets, which offer limited applicability for everyday usage. With the aim to bring brainwave authentication and its benefits closer to real world deployment, we investigate brain biometrics with consumer devices. We conduct a comprehensive measurement experiment and user study that compare five authentication tasks on a user sample up to 10 times larger than those from previous studies, introducing three novel techniques based on cognitive semantic processing. Furthermore, we apply our analysis on high-quality open brainwave data obtained with a medical-grade headset, to assess the differences. We investigate both the performance, security, and usability of the different options and use this evidence to elicit design and research recommendations. Our results show that it is possible to achieve Equal Error Rates as low as 7.2% (a reduction between 68–72% with respect to existing approaches) based on brain responses to images with current inexpensive technology. We show that the common practice of testing authentication systems only with known attacker data is unrealistic and may lead to overly optimistic evaluations. With regard to adoption, users call for simpler devices, faster authentication, and better privacy.
{"title":"Performance and Usability Evaluation of Brainwave Authentication Techniques with Consumer Devices","authors":"Patricia Arias-Cabarcos, Matin Fallahi, Thilo Habrich, Karen Schulze, Christian Becker, Thorsten Strufe","doi":"10.1145/3579356","DOIUrl":"https://doi.org/10.1145/3579356","url":null,"abstract":"Brainwaves have demonstrated to be unique enough across individuals to be useful as biometrics. They also provide promising advantages over traditional means of authentication, such as resistance to external observability, revocability, and intrinsic liveness detection. However, most of the research so far has been conducted with expensive, bulky, medical-grade helmets, which offer limited applicability for everyday usage. With the aim to bring brainwave authentication and its benefits closer to real world deployment, we investigate brain biometrics with consumer devices. We conduct a comprehensive measurement experiment and user study that compare five authentication tasks on a user sample up to 10 times larger than those from previous studies, introducing three novel techniques based on cognitive semantic processing. Furthermore, we apply our analysis on high-quality open brainwave data obtained with a medical-grade headset, to assess the differences. We investigate both the performance, security, and usability of the different options and use this evidence to elicit design and research recommendations. Our results show that it is possible to achieve Equal Error Rates as low as 7.2% (a reduction between 68–72% with respect to existing approaches) based on brain responses to images with current inexpensive technology. We show that the common practice of testing authentication systems only with known attacker data is unrealistic and may lead to overly optimistic evaluations. With regard to adoption, users call for simpler devices, faster authentication, and better privacy.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"26 1","pages":"1 - 36"},"PeriodicalIF":2.3,"publicationDate":"2023-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42165086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The unprecedented growth in mobile systems has transformed the way we approach everyday computing. Unfortunately, the emergence of a sophisticated type of malware known as ransomware poses a great threat to consumers of this technology. Traditional research on mobile malware detection has focused on approaches that rely on analyzing bytecode for uncovering malicious apps. However, cybercriminals can bypass such methods by embedding malware directly in native machine code, making traditional methods inadequate. Another challenge that detection solutions face is scalability. The sheer number of malware variants released every year makes it difficult for solutions to efficiently scale their coverage. To address these concerns, this work presents RansomShield, an energy-efficient solution that leverages CNNs to detect ransomware. We evaluate CNN architectures that have been known to perform well on computer vision tasks and examine their suitability for ransomware detection. We show that systematically converting native instructions from Android apps into images using space-filling curve visualization techniques enable CNNs to reliably detect ransomware with high accuracy. We characterize the robustness of this approach across ARM and x86 architectures and demonstrate the effectiveness of this solution across heterogeneous platforms including smartphones and chromebooks. We evaluate the suitability of different models for mobile systems by comparing their energy demands using different platforms. In addition, we present a CNN introspection framework that determines the important features that are needed for ransomware detection. Finally, we evaluate the robustness of this solution against adversarial machine learning (AML) attacks using state-of-the-art Android malware dataset.
{"title":"RansomShield: A Visualization Approach to Defending Mobile Systems Against Ransomware","authors":"Nada Lachtar, Duha Ibdah, Hamza Khan, Anys Bacha","doi":"10.1145/3579822","DOIUrl":"https://doi.org/10.1145/3579822","url":null,"abstract":"The unprecedented growth in mobile systems has transformed the way we approach everyday computing. Unfortunately, the emergence of a sophisticated type of malware known as ransomware poses a great threat to consumers of this technology. Traditional research on mobile malware detection has focused on approaches that rely on analyzing bytecode for uncovering malicious apps. However, cybercriminals can bypass such methods by embedding malware directly in native machine code, making traditional methods inadequate. Another challenge that detection solutions face is scalability. The sheer number of malware variants released every year makes it difficult for solutions to efficiently scale their coverage. To address these concerns, this work presents RansomShield, an energy-efficient solution that leverages CNNs to detect ransomware. We evaluate CNN architectures that have been known to perform well on computer vision tasks and examine their suitability for ransomware detection. We show that systematically converting native instructions from Android apps into images using space-filling curve visualization techniques enable CNNs to reliably detect ransomware with high accuracy. We characterize the robustness of this approach across ARM and x86 architectures and demonstrate the effectiveness of this solution across heterogeneous platforms including smartphones and chromebooks. We evaluate the suitability of different models for mobile systems by comparing their energy demands using different platforms. In addition, we present a CNN introspection framework that determines the important features that are needed for ransomware detection. Finally, we evaluate the robustness of this solution against adversarial machine learning (AML) attacks using state-of-the-art Android malware dataset.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"26 1","pages":"1 - 30"},"PeriodicalIF":2.3,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42844301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-14DOI: https://dl.acm.org/doi/10.1145/3568020
Farzana Ahamed Bhuiyan, Akond Rahman
Adversarial attacks against supervised learning algorithms, which necessitates the application of logging while using supervised learning algorithms in software projects. Logging enables practitioners to conduct postmortem analysis, which can be helpful to diagnose any conducted attacks. We conduct an empirical study to identify and characterize log-related coding patterns, i.e., recurring coding patterns that can be leveraged to conduct adversarial attacks and needs to be logged. A list of log-related coding patterns can guide practitioners on what to log while using supervised learning algorithms in software projects.
We apply qualitative analysis on 3,004 Python files used to implement 103 supervised learning-based software projects. We identify a list of 54 log-related coding patterns that map to 6 attacks related to supervised learning algorithms. Using Log Assistant to conductPostmortems forSupervisedLearning (LOPSUL), we quantify the frequency of the identified log-related coding patterns with 278 open source software projects that use supervised learning. We observe log-related coding patterns to appear for 22% of the analyzed files, where training data forensics is the most frequently occurring category.
{"title":"Log-related Coding Patterns to Conduct Postmortems of Attacks in Supervised Learning-based Projects","authors":"Farzana Ahamed Bhuiyan, Akond Rahman","doi":"https://dl.acm.org/doi/10.1145/3568020","DOIUrl":"https://doi.org/https://dl.acm.org/doi/10.1145/3568020","url":null,"abstract":"<p>Adversarial attacks against supervised learning algorithms, which necessitates the application of logging while using supervised learning algorithms in software projects. Logging enables practitioners to conduct postmortem analysis, which can be helpful to diagnose any conducted attacks. We conduct an empirical study to identify and characterize log-related coding patterns, i.e., recurring coding patterns that can be leveraged to conduct adversarial attacks and needs to be logged. A list of log-related coding patterns can guide practitioners on what to log while using supervised learning algorithms in software projects. </p><p>We apply qualitative analysis on 3,004 Python files used to implement 103 supervised learning-based software projects. We identify a list of 54 log-related coding patterns that map to 6 attacks related to supervised learning algorithms. Using <i><b>Lo</b><i>g Assistant to conduct</i><b>P</b><i>ostmortems for</i><b>Su</b><i>pervised</i><b>L</b><i>earning</i> (<b>LOPSUL</b></i>), we quantify the frequency of the identified log-related coding patterns with 278 open source software projects that use supervised learning. We observe log-related coding patterns to appear for 22% of the analyzed files, where training data forensics is the most frequently occurring category.</p>","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":"11 1","pages":""},"PeriodicalIF":2.3,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138540626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Adversarial attacks against supervised learninga algorithms, which necessitates the application of logging while using supervised learning algorithms in software projects. Logging enables practitioners to conduct postmortem analysis, which can be helpful to diagnose any conducted attacks. We conduct an empirical study to identify and characterize log-related coding patterns, i.e., recurring coding patterns that can be leveraged to conduct adversarial attacks and needs to be logged. A list of log-related coding patterns can guide practitioners on what to log while using supervised learning algorithms in software projects. We apply qualitative analysis on 3,004 Python files used to implement 103 supervised learning-based software projects. We identify a list of 54 log-related coding patterns that map to six attacks related to supervised learning algorithms. Using Log Assistant to conduct Postmortems for Supervised Learning (LOPSUL), we quantify the frequency of the identified log-related coding patterns with 278 open-source software projects that use supervised learning. We observe log-related coding patterns to appear for 22% of the analyzed files, where training data forensics is the most frequently occurring category.
{"title":"Log-related Coding Patterns to Conduct Postmortems of Attacks in Supervised Learning-based Projects","authors":"Farzana Ahamed Bhuiyan, A. Rahman","doi":"10.1145/3568020","DOIUrl":"https://doi.org/10.1145/3568020","url":null,"abstract":"Adversarial attacks against supervised learninga algorithms, which necessitates the application of logging while using supervised learning algorithms in software projects. Logging enables practitioners to conduct postmortem analysis, which can be helpful to diagnose any conducted attacks. We conduct an empirical study to identify and characterize log-related coding patterns, i.e., recurring coding patterns that can be leveraged to conduct adversarial attacks and needs to be logged. A list of log-related coding patterns can guide practitioners on what to log while using supervised learning algorithms in software projects. We apply qualitative analysis on 3,004 Python files used to implement 103 supervised learning-based software projects. We identify a list of 54 log-related coding patterns that map to six attacks related to supervised learning algorithms. Using Log Assistant to conduct Postmortems for Supervised Learning (LOPSUL), we quantify the frequency of the identified log-related coding patterns with 278 open-source software projects that use supervised learning. We observe log-related coding patterns to appear for 22% of the analyzed files, where training data forensics is the most frequently occurring category.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":" ","pages":"1 - 24"},"PeriodicalIF":2.3,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48834916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Exciting recent advances in genome sequencing, coupled with greatly reduced storage and computation costs, make genomic testing increasingly accessible to individuals. Already today, one’s digitized DNA can be easily obtained from a sequencing lab and later used to conduct numerous tests by engaging with a testing facility. Due to the inherent sensitivity of genetic material and the often-proprietary nature of genomic tests, privacy is a natural and crucial issue. While genomic privacy received a great deal of attention within and outside the research community, genomic security has not been sufficiently studied. This is surprising since the usage of fake or altered genomes can have grave consequences, such as erroneous drug prescriptions and genetic test outcomes. Unfortunately, in the genomic domain, privacy and security (as often happens) are at odds with each other. In this article, we attempt to reconcile security with privacy in genomic testing by designing a novel technique for a secure and private genomic range query protocol between a genomic testing facility and an individual user. The proposed technique ensures authenticity and completeness of user-supplied genomic material while maintaining its privacy by releasing only the minimum thereof. To confirm its broad usability, we show how to apply the proposed technique to a previously proposed genomic private substring matching protocol. Experiments show that the proposed technique offers good performance and is quite practical. Furthermore, we generalize the genomic range query problem to sparse integer sets and discuss potential use cases.
{"title":"Balancing Security and Privacy in Genomic Range Queries","authors":"Seoyeon Hwang, Ercan Ozturk, G. Tsudik","doi":"10.1145/3575796","DOIUrl":"https://doi.org/10.1145/3575796","url":null,"abstract":"Exciting recent advances in genome sequencing, coupled with greatly reduced storage and computation costs, make genomic testing increasingly accessible to individuals. Already today, one’s digitized DNA can be easily obtained from a sequencing lab and later used to conduct numerous tests by engaging with a testing facility. Due to the inherent sensitivity of genetic material and the often-proprietary nature of genomic tests, privacy is a natural and crucial issue. While genomic privacy received a great deal of attention within and outside the research community, genomic security has not been sufficiently studied. This is surprising since the usage of fake or altered genomes can have grave consequences, such as erroneous drug prescriptions and genetic test outcomes. Unfortunately, in the genomic domain, privacy and security (as often happens) are at odds with each other. In this article, we attempt to reconcile security with privacy in genomic testing by designing a novel technique for a secure and private genomic range query protocol between a genomic testing facility and an individual user. The proposed technique ensures authenticity and completeness of user-supplied genomic material while maintaining its privacy by releasing only the minimum thereof. To confirm its broad usability, we show how to apply the proposed technique to a previously proposed genomic private substring matching protocol. Experiments show that the proposed technique offers good performance and is quite practical. Furthermore, we generalize the genomic range query problem to sparse integer sets and discuss potential use cases.","PeriodicalId":56050,"journal":{"name":"ACM Transactions on Privacy and Security","volume":" ","pages":"1 - 28"},"PeriodicalIF":2.3,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42373387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号