Dokyung Song, Julian Lettner, Prabhu Rajasekaran, Yeoul Na, Stijn Volckaert, Per Larsen, M. Franz
The C and C++ programming languages are notoriously insecure yet remain indispensable. Developers therefore resort to a multi-pronged approach to find security issues before adversaries. These include manual, static, and dynamic program analysis. Dynamic bug finding tools—henceforth "sanitizers"—can find bugs that elude other types of analysis because they observe the actual execution of a program, and can therefore directly observe incorrect program behavior as it happens. A vast number of sanitizers have been prototyped by academics and refined by practitioners. We provide a systematic overview of sanitizers with an emphasis on their role in finding security issues. Specifically, we taxonomize the available tools and the security vulnerabilities they cover, describe their performance and compatibility properties, and highlight various trade-offs.
{"title":"SoK: Sanitizing for Security","authors":"Dokyung Song, Julian Lettner, Prabhu Rajasekaran, Yeoul Na, Stijn Volckaert, Per Larsen, M. Franz","doi":"10.1109/SP.2019.00010","DOIUrl":"https://doi.org/10.1109/SP.2019.00010","url":null,"abstract":"The C and C++ programming languages are notoriously insecure yet remain indispensable. Developers therefore resort to a multi-pronged approach to find security issues before adversaries. These include manual, static, and dynamic program analysis. Dynamic bug finding tools—henceforth \"sanitizers\"—can find bugs that elude other types of analysis because they observe the actual execution of a program, and can therefore directly observe incorrect program behavior as it happens. A vast number of sanitizers have been prototyped by academics and refined by practitioners. We provide a systematic overview of sanitizers with an emphasis on their role in finding security issues. Specifically, we taxonomize the available tools and the security vulnerabilities they cover, describe their performance and compatibility properties, and highlight various trade-offs.","PeriodicalId":272713,"journal":{"name":"2019 IEEE Symposium on Security and Privacy (SP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128630226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Luca Melis, Congzheng Song, Emiliano De Cristofaro, Vitaly Shmatikov
Collaborative machine learning and related techniques such as federated learning allow multiple participants, each with his own training dataset, to build a joint model by training locally and periodically exchanging model updates. We demonstrate that these updates leak unintended information about participants' training data and develop passive and active inference attacks to exploit this leakage. First, we show that an adversarial participant can infer the presence of exact data points -- for example, specific locations -- in others' training data (i.e., membership inference). Then, we show how this adversary can infer properties that hold only for a subset of the training data and are independent of the properties that the joint model aims to capture. For example, he can infer when a specific person first appears in the photos used to train a binary gender classifier. We evaluate our attacks on a variety of tasks, datasets, and learning configurations, analyze their limitations, and discuss possible defenses.
{"title":"Exploiting Unintended Feature Leakage in Collaborative Learning","authors":"Luca Melis, Congzheng Song, Emiliano De Cristofaro, Vitaly Shmatikov","doi":"10.1109/SP.2019.00029","DOIUrl":"https://doi.org/10.1109/SP.2019.00029","url":null,"abstract":"Collaborative machine learning and related techniques such as federated learning allow multiple participants, each with his own training dataset, to build a joint model by training locally and periodically exchanging model updates. We demonstrate that these updates leak unintended information about participants' training data and develop passive and active inference attacks to exploit this leakage. First, we show that an adversarial participant can infer the presence of exact data points -- for example, specific locations -- in others' training data (i.e., membership inference). Then, we show how this adversary can infer properties that hold only for a subset of the training data and are independent of the properties that the joint model aims to capture. For example, he can infer when a specific person first appears in the photos used to train a binary gender classifier. We evaluate our attacks on a variety of tasks, datasets, and learning configurations, analyze their limitations, and discuss possible defenses.","PeriodicalId":272713,"journal":{"name":"2019 IEEE Symposium on Security and Privacy (SP)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132676543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mathias Lécuyer, Vaggelis Atlidakis, Roxana Geambasu, Daniel J. Hsu, S. Jana
Adversarial examples that fool machine learning models, particularly deep neural networks, have been a topic of intense research interest, with attacks and defenses being developed in a tight back-and-forth. Most past defenses are best effort and have been shown to be vulnerable to sophisticated attacks. Recently a set of certified defenses have been introduced, which provide guarantees of robustness to norm-bounded attacks. However these defenses either do not scale to large datasets or are limited in the types of models they can support. This paper presents the first certified defense that both scales to large networks and datasets (such as Google’s Inception network for ImageNet) and applies broadly to arbitrary model types. Our defense, called PixelDP, is based on a novel connection between robustness against adversarial examples and differential privacy, a cryptographically-inspired privacy formalism, that provides a rigorous, generic, and flexible foundation for defense.
{"title":"Certified Robustness to Adversarial Examples with Differential Privacy","authors":"Mathias Lécuyer, Vaggelis Atlidakis, Roxana Geambasu, Daniel J. Hsu, S. Jana","doi":"10.1109/SP.2019.00044","DOIUrl":"https://doi.org/10.1109/SP.2019.00044","url":null,"abstract":"Adversarial examples that fool machine learning models, particularly deep neural networks, have been a topic of intense research interest, with attacks and defenses being developed in a tight back-and-forth. Most past defenses are best effort and have been shown to be vulnerable to sophisticated attacks. Recently a set of certified defenses have been introduced, which provide guarantees of robustness to norm-bounded attacks. However these defenses either do not scale to large datasets or are limited in the types of models they can support. This paper presents the first certified defense that both scales to large networks and datasets (such as Google’s Inception network for ImageNet) and applies broadly to arbitrary model types. Our defense, called PixelDP, is based on a novel connection between robustness against adversarial examples and differential privacy, a cryptographically-inspired privacy formalism, that provides a rigorous, generic, and flexible foundation for defense.","PeriodicalId":272713,"journal":{"name":"2019 IEEE Symposium on Security and Privacy (SP)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128073467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
P. Kocher, Daniel Genkin, D. Gruss, Werner Haas, Michael Hamburg, Moritz Lipp, S. Mangard, Thomas Prescher, Michael Schwarz, Y. Yarom
Modern processors use branch prediction and speculative execution to maximize performance. For example, if the destination of a branch depends on a memory value that is in the process of being read, CPUs will try to guess the destination and attempt to execute ahead. When the memory value finally arrives, the CPU either discards or commits the speculative computation. Speculative logic is unfaithful in how it executes, can access the victim's memory and registers, and can perform operations with measurable side effects. Spectre attacks involve inducing a victim to speculatively perform operations that would not occur during correct program execution and which leak the victim's confidential information via a side channel to the adversary. This paper describes practical attacks that combine methodology from side channel attacks, fault attacks, and return-oriented programming that can read arbitrary memory from the victim's process. More broadly, the paper shows that speculative execution implementations violate the security assumptions underpinning numerous software security mechanisms, including operating system process separation, containerization, just-in-time (JIT) compilation, and countermeasures to cache timing and side-channel attacks. These attacks represent a serious threat to actual systems since vulnerable speculative execution capabilities are found in microprocessors from Intel, AMD, and ARM that are used in billions of devices. While makeshift processor-specific countermeasures are possible in some cases, sound solutions will require fixes to processor designs as well as updates to instruction set architectures (ISAs) to give hardware architects and software developers a common understanding as to what computation state CPU implementations are (and are not) permitted to leak.
{"title":"Spectre Attacks: Exploiting Speculative Execution","authors":"P. Kocher, Daniel Genkin, D. Gruss, Werner Haas, Michael Hamburg, Moritz Lipp, S. Mangard, Thomas Prescher, Michael Schwarz, Y. Yarom","doi":"10.1145/3399742","DOIUrl":"https://doi.org/10.1145/3399742","url":null,"abstract":"Modern processors use branch prediction and speculative execution to maximize performance. For example, if the destination of a branch depends on a memory value that is in the process of being read, CPUs will try to guess the destination and attempt to execute ahead. When the memory value finally arrives, the CPU either discards or commits the speculative computation. Speculative logic is unfaithful in how it executes, can access the victim's memory and registers, and can perform operations with measurable side effects. Spectre attacks involve inducing a victim to speculatively perform operations that would not occur during correct program execution and which leak the victim's confidential information via a side channel to the adversary. This paper describes practical attacks that combine methodology from side channel attacks, fault attacks, and return-oriented programming that can read arbitrary memory from the victim's process. More broadly, the paper shows that speculative execution implementations violate the security assumptions underpinning numerous software security mechanisms, including operating system process separation, containerization, just-in-time (JIT) compilation, and countermeasures to cache timing and side-channel attacks. These attacks represent a serious threat to actual systems since vulnerable speculative execution capabilities are found in microprocessors from Intel, AMD, and ARM that are used in billions of devices. While makeshift processor-specific countermeasures are possible in some cases, sound solutions will require fixes to processor designs as well as updates to instruction set architectures (ISAs) to give hardware architects and software developers a common understanding as to what computation state CPU implementations are (and are not) permitted to leak.","PeriodicalId":272713,"journal":{"name":"2019 IEEE Symposium on Security and Privacy (SP)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128451462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Manu Drijvers, Kasra Edalatnejad, B. Ford, Eike Kiltz, J. Loss, G. Neven, Igors Stepanovs
A multi-signature scheme allows a group of signers to collaboratively sign a message, creating a single signature that convinces a verifier that every individual signer approved the message. The increased interest in technologies to decentralize trust has triggered the proposal of highly efficient two-round Schnorr-based multi-signature schemes designed to scale up to thousands of signers, namely BCJ by Bagherzandi et al. (CCS 2008), MWLD by Ma et al. (DCC 2010), CoSi by Syta et al. (S&P 2016), and MuSig by Maxwell et al. (ePrint 2018). In this work, we point out serious security issues in all currently known two-round multi-signature schemes (without pairings). First, we prove that none of the schemes can be proved secure without radically departing from currently known techniques. Namely, we show that if the one-more discrete-logarithm problem is hard, then no algebraic reduction exists that proves any of these schemes secure under the discrete-logarithm or one-more discrete-logarithm problem. We point out subtle flaws in the published security proofs of the above schemes (except CoSi, which was not proved secure) to clarify the contradiction between our result and the existing proofs. Next, we describe practical sub-exponential attacks on all schemes, providing further evidence to their insecurity. Being left without two-round multi-signature schemes, we present mBCJ, a variant of the BCJ scheme that we prove secure under the discrete-logarithm assumption in the random-oracle model. Our experiments show that mBCJ barely affects scalability compared to CoSi, allowing 16384 signers to collaboratively sign a message in about 2 seconds, making it a highly practical and provably secure alternative for large-scale deployments.
{"title":"On the Security of Two-Round Multi-Signatures","authors":"Manu Drijvers, Kasra Edalatnejad, B. Ford, Eike Kiltz, J. Loss, G. Neven, Igors Stepanovs","doi":"10.1109/SP.2019.00050","DOIUrl":"https://doi.org/10.1109/SP.2019.00050","url":null,"abstract":"A multi-signature scheme allows a group of signers to collaboratively sign a message, creating a single signature that convinces a verifier that every individual signer approved the message. The increased interest in technologies to decentralize trust has triggered the proposal of highly efficient two-round Schnorr-based multi-signature schemes designed to scale up to thousands of signers, namely BCJ by Bagherzandi et al. (CCS 2008), MWLD by Ma et al. (DCC 2010), CoSi by Syta et al. (S&P 2016), and MuSig by Maxwell et al. (ePrint 2018). In this work, we point out serious security issues in all currently known two-round multi-signature schemes (without pairings). First, we prove that none of the schemes can be proved secure without radically departing from currently known techniques. Namely, we show that if the one-more discrete-logarithm problem is hard, then no algebraic reduction exists that proves any of these schemes secure under the discrete-logarithm or one-more discrete-logarithm problem. We point out subtle flaws in the published security proofs of the above schemes (except CoSi, which was not proved secure) to clarify the contradiction between our result and the existing proofs. Next, we describe practical sub-exponential attacks on all schemes, providing further evidence to their insecurity. Being left without two-round multi-signature schemes, we present mBCJ, a variant of the BCJ scheme that we prove secure under the discrete-logarithm assumption in the random-oracle model. Our experiments show that mBCJ barely affects scalability compared to CoSi, allowing 16384 signers to collaboratively sign a message in about 2 seconds, making it a highly practical and provably secure alternative for large-scale deployments.","PeriodicalId":272713,"journal":{"name":"2019 IEEE Symposium on Security and Privacy (SP)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124550351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：