Filters








87 Hits in 5.4 sec

Breaking certified defenses: Semantic adversarial examples with spoofed robustness certificates [article]

Amin Ghiasi, Ali Shafahi, Tom Goldstein
2020 arXiv   pre-print
The proposed "Shadow Attack" causes certifiably robust networks to mislabel an image and simultaneously produce a "spoofed" certificate of robustness.  ...  In addition to labeling an image, certified classifiers produce (when possible) a certificate guaranteeing that the input image is not an ℓ_p-bounded adversarial example.  ...  CONCLUSION We demonstrate that it is possible to produce adversarial examples with "spoofed" certified robustness by using large-norm perturbations.  ... 
arXiv:2003.08937v1 fatcat:gnscxyhpc5ckdj2xl54yfg227a

Advances in adversarial attacks and defenses in computer vision: A survey [article]

Naveed Akhtar, Ajmal Mian, Navid Kardan, Mubarak Shah
2021 arXiv   pre-print
In [2], we reviewed the contributions made by the computer vision community in adversarial attacks on deep learning (and their defenses) until the advent of year 2018.  ...  However, it is now known that DL is vulnerable to adversarial attacks that can manipulate its predictions by introducing visually imperceptible perturbations in images and videos.  ...  Through generating the perturbation outside the certified p bounds, their method produces a "spoofed" certificate, which results in visually imperceptible adversarial perturbations to break the defense  ... 
arXiv:2108.00401v2 fatcat:23gw74oj6bblnpbpeacpg3hq5y

Adversarial Attacks and Defenses in Images, Graphs and Text: A Review [article]

Han Xu, Yao Ma, Haochen Liu, Debayan Deb, Hui Liu, Jiliang Tang, Anil K. Jain
2019 arXiv   pre-print
In this survey, we review the state of the art algorithms for generating adversarial examples and the countermeasures against adversarial examples, for the three popular data types, i.e., images, graphs  ...  However, the existence of adversarial examples has raised concerns about applying deep learning to safety-critical applications.  ...  samples augmented with adversarial examples, they increase the robustness against adversarial examples generated by FGSM.  ... 
arXiv:1909.08072v2 fatcat:i3han24f3fdgpop45t4pmxcdtm

The Security Reference Architecture for Blockchains: Towards a Standardized Model for Studying Vulnerabilities, Threats, and Defenses [article]

Ivan Homoliak, Sarad Venugopalan, Qingze Hum, Daniel Reijsbergen, Richard Schumi, Pawel Szalachowski
2020 arXiv   pre-print
Finally, we provide designers of blockchain platforms and applications with a design methodology following the model of SRA and its hierarchy.  ...  Nevertheless, the blockchain "does not understand" the semantics of data that are timestamped, and thus it cannot vet or certify them.  ...  Semantic Bugs. We deal with semantic bugs only at the level of the RSM layer as part of the smart contract code (see Section VII-B).  ... 
arXiv:1910.09775v2 fatcat:xvxnekjhtbbt7dc4b3kxy3m6xi

The Security Reference Architecture for Blockchains: Towards a Standardized Model for Studying Vulnerabilities, Threats, and Defenses

Ivan Homoliak, Sarad Venugopalan, Daniel Reijsbergen, Qingze Hum, Richard Schumi, Pawel Szalachowski
2020 IEEE Communications Surveys and Tutorials  
Finally, we provide designers of blockchain platforms and applications with a design methodology following the model of SRA and its hierarchy.  ...  Nevertheless, the blockchain "does not understand" the semantics of data that are timestamped, and thus it cannot vet or certify them.  ...  ) Semantic Bugs: We deal with semantic bugs only at the level of the RSM layer as part of the smart contracts (see Section VII-B).  ... 
doi:10.1109/comst.2020.3033665 fatcat:hcrsvvkb2ffmbdlspshy66wg2y

Defense Advanced Research Projects Agency (Darpa) Fiscal Year 2015 Budget Estimates

Department Of Defense Comptroller's Office
2014 Zenodo  
Drug Administration (FDA), and the Defense Advanced Research Projects Agency (DARPA).  ...  The Defense Advanced Research Projects Agency (DARPA) FY2015 amounted to $2.9 billion in the President's request to support high-risk, high-reward research.  ...  Techniques will be developed to disrupt and degrade adversary information systems and network operations, with particular interest in techniques likely to be robust to adversary countermeasure strategies  ... 
doi:10.5281/zenodo.1215345 fatcat:fjzhmynqjbaafk67q2ckcblj2m

Measuring Adversarial Robustness using a Voronoi-Epsilon Adversary [article]

Hyeongji Kim, Pekka Parviainen, Ketil Malde
2021 arXiv   pre-print
As ϵ gets large, the adversary may use real data points from other classes as adversarial examples. We propose a Voronoi-epsilon adversary which is constrained both by Voronoi cells and by ϵ-balls.  ...  Finally, we show that a nearest neighbor classifier is the maximally robust classifier against the proposed adversary on the training data.  ...  Breaking certified defenses: Semantic adversarial examples with spoofed robustness certificates. In International Conference on Learning Repre- sentations, 2019.  ... 
arXiv:2005.02540v3 fatcat:x4pjxrf4mffdtezlxibf3uc2ra

Biometrics: Trust, but Verify [article]

Anil K. Jain, Debayan Deb, Joshua J. Engelsma
2021 arXiv   pre-print
Some of these problems include: i) questions related to system recognition performance, ii) security (spoof attacks, adversarial attacks, template reconstruction attacks and demographic information leakage  ...  For example, the German Chaos Computer Club demonstrated with ease the breaking of Apple's TouchID already in 2013 8 .  ...  A common defense strategy, namely adversarial training, is to re-train the classifier we wish to defend with perturbation attacks [148, [154] [155] [156] [157] .  ... 
arXiv:2105.06625v2 fatcat:gnii3qxufzevzlkfimhmmwof7i

Pretty-Bad-Proxy: An Overlooked Adversary in Browsers' HTTPS Deployments

Shuo Chen, Ziqing Mao, Yi-Min Wang, Ming Zhang
2009 2009 30th IEEE Symposium on Security and Privacy  
It attempts to break the end-to-end security guarantees of HTTPS without breaking any cryptographic scheme.  ...  The protocol itself has been rigorously designed and evaluated by assuming the network as an adversary.  ...  Our communications with Adam Barth and Collin Jackson helped improve the paper.  ... 
doi:10.1109/sp.2009.12 dblp:conf/sp/ChenMWZ09 fatcat:kqnek57ldjh4hmwv7zosmepcxy

VERDICT: A Language and Framework for Engineering Cyber Resilient and Safe System

Baoluo Meng, Daniel Larraz, Kit Siu, Abha Moitra, John Interrante, William Smith, Saswata Paul, Daniel Prince, Heber Herencia-Zapana, M. Fareed Arif, Moosa Yahyazadeh, Vidhya Tekken Valapil (+3 others)
2021 Systems  
It can also synthesize optimal defense solutions—with respect to implementation costs. In addition, ACFG assembles MBAAS evidence into goal structuring notation for certification purposes.  ...  MBAAS performs probabilistic analyses, suggests defenses to mitigate attacks, and generates attack-defense trees and fault trees as evidence of resiliency and safety.  ...  Even if the system uses a robust GPS sensor to resist spoofing attacks, due to the lack of behavioral details in the architecture we would not be able to rule out such attacks.  ... 
doi:10.3390/systems9010018 fatcat:73d2wytatnddvlaihhce5dkoba

Securing Public Safety Communications on Commercial and Tactical 5G Networks: A Survey and Future Research Directions

Jani Suomalainen, Jukka Julku, Mikko Vehkapera, Harri Posti
2021 IEEE Open Journal of the Communications Society  
This transition from closed and dedicated infrastructure to hybrid architecture will expand the threat surface and expose mission-critical applications and sensitive information to cyber and physical adversaries  ...  discuss future research directions highlighting the need for enablers to automate security of rapid deployments, for military-grade cost-effective customizations of commercial network services to ensure robustness  ...  Typical PEP implementations also break the end-to-end semantics of IP connections and prohibit the use of IPsec.  ... 
doi:10.1109/ojcoms.2021.3093529 fatcat:5d66gh3b2vcinddaml44oznpqe

The Threat of Adversarial Attacks on Machine Learning in Network Security – A Survey [article]

Olakunle Ibitoye, Rana Abou-Khamis, Ashraf Matrawy, M. Omair Shafiq
2020 arXiv   pre-print
We then analyze the various defenses against adversarial attacks on machine learning-based network security applications.  ...  In what could be considered an arms race between attackers and defenders, adversaries constantly probe machine learning systems with inputs which are explicitly designed to bypass the system and induce  ...  [87] which is a new method for attacking systems that rely on certificates and fool certified robust networks to assign the wrong label to an image and produce a spoofed secure robustness certificate  ... 
arXiv:1911.02621v2 fatcat:p7mgj65wavee3op6as5lufwj3q

Security and Privacy Issues in Wireless Mesh Networks: A Survey [article]

Jaydip Sen
2013 arXiv   pre-print
security and user privacy in WMNs, this chapter provides a comprehensive overview of various possible attacks on different layers of the communication protocol stack for WMNs and their corresponding defense  ...  Comparative analyses are also presented on the security schemes with regards to the cryptographic schemes used, key management strategies deployed, use of any trusted third party, computation and communication  ...  In TCAODV, each node has its public key certified by a certificate authority (CA) and the public key certificate of each node is stored within a trusted root.  ... 
arXiv:1302.0939v1 fatcat:4bxooy2dubeopa3qbcyg5yvi3y

Securing Vehicle-to-Everything (V2X) Communication Platforms [article]

Monowar Hasan, Sibin Mohan, Takayuki Shimizu, Hongsheng Lu
2020 arXiv   pre-print
We then identified semantic gaps of existing security solutions and outline possible open issues.  ...  We also review main security/privacy issues, current standardization activities and existing defense mechanisms proposed within the V2X domain.  ...  ., root certificate authority (RCA), that provides services to authenticate the identity of entities; (b) a registration authority certified by an RCA that issues certificates for specific uses permitted  ... 
arXiv:2003.07191v1 fatcat:rtyelz5ro5agdf5szamdn4tsvm

Securing Vehicle-to-Everything (V2X) Communication Platforms

Monowar Hasan, Sibin Mohan, Takayuki Shimizu, Hongsheng Lu
2020 IEEE Transactions on Intelligent Vehicles  
We then identified semantic gaps of existing security solutions and outline possible open issues.  ...  We also review main security/privacy issues, current standardization activities and existing defense mechanisms proposed within the V2X domain.  ...  ., root certificate authority (RCA), that provides services to authenticate the identity of entities; (b) a registration authority certified by an RCA that issues certificates for specific uses permitted  ... 
doi:10.1109/tiv.2020.2987430 fatcat:egw4hik2onfarc6nsjsng2r6ei
« Previous Showing results 1 — 15 out of 87 results