Filters








4,923 Hits in 4.4 sec

Explainability-Aware One Point Attack for Point Cloud Neural Networks [article]

Hanxiao Tan, Helena Kotthaus
2022 arXiv   pre-print
Finally, we discuss how our approaches facilitate the explainability study for point cloud networks.  ...  reliability of point cloud networks by adversarial attacks.  ...  OPA on 2D image neural network For a relatively fair comparison as a reference, we extend our OPA to 2D image neural networks for a rough comparison of its sensitivity to critical points with that of 3D  ... 
arXiv:2110.04158v3 fatcat:aeq4v45tzvfslknic7otxarla4

Dynamics-aware Adversarial Attack of 3D Sparse Convolution Network [article]

An Tao and Yueqi Duan and He Wang and Ziyi Wu and Pengliang Ji and Haowen Sun and Jie Zhou and Jiwen Lu
2021 arXiv   pre-print
In this paper, we investigate the dynamics-aware adversarial attack problem in deep neural networks.  ...  Compared with the dynamic-unaware methods, LGM achieves about 20% lower mIoU averagely on the ScanNet and S3DIS datasets. LGM also outperforms the recent point cloud attacks.  ...  Instead, our dynamics-aware attack achieves remarkably lower mIoU on the presented point cloud scene.  ... 
arXiv:2112.09428v1 fatcat:qz4u6vrm5zfstd6z6u6yeqgh6u

User configurable 3D object regeneration for spatial privacy [article]

Arpit Nama, Amaya Dharmasiri, Kanchana Thilakarathna, Albert Zomaya, Jaybie Agullo de Guzman
2021 arXiv   pre-print
Thus, in this work, we demonstrate how we can leverage 3D object regeneration for preserving privacy of 3D point clouds.  ...  That is, we employ an intermediary layer of protection to transform the 3D point cloud before providing it to the third-party applications.  ...  PointNet was the first to use deep learning, i.e. deep neural network (DNN), for 3D classification with point clouds as input [27] .  ... 
arXiv:2108.08273v1 fatcat:hpbbdoaw3rfaxcc5z4b5pityvu

DUP-Net: Denoiser and Upsampler Network for 3D Adversarial Point Clouds Defense [article]

Hang Zhou, Kejiang Chen, Weiming Zhang, Han Fang, Wenbo Zhou and Nenghai Yu
2019 arXiv   pre-print
Third, the upsampler network can be trained on a small dataset and defends well against adversarial attacks generated from other point cloud datasets.  ...  We propose a Denoiser and UPsampler Network (DUP-Net) structure as defenses for 3D adversarial point cloud classification, where the two modules reconstruct surface smoothness by dropping or adding points  ...  The tendency of three curves can be explained below: the attacks search the entire point cloud space for adversarial perturbations without regarding for the location of the point cloud content.  ... 
arXiv:1812.11017v2 fatcat:xnilxyqtkvdhzelpvpjc37ucbm

Privacy-Preserving Cloud Computing: Ecosystem, Life Cycle, Layered Architecture and Future Roadmap [article]

Saeed Ahmadi
2022 arXiv   pre-print
This survey paper on privacy-preserving cloud computing can help pave the way for future research in related areas.  ...  Likewise, privacy in cloud computing is important because it ensures that the integrity of data stored on the cloud maintains intact.  ...  The authors in [357] have argued supervised and unsupervised machine learning capability through neural networks on encrypted data similar to [358] that for privacy-preserving deep neural networks,  ... 
arXiv:2204.11120v1 fatcat:tx75pckegjgqxg6tiibptiazbi

INTRUSION DETECTION SYSTEM IN SECURE SHELL TRAFFIC IN CLOUD ENVIRONMENT

Mehdi Barati, Azizol Abdullah, NurIzura Udzir, Mostafa Behzadi, Ramlan Mahmod, Norwati Mustapha
2014 Journal of Computer Science  
Brute force attack is launched by implementing a client-server SSH model in a private Cloud environment and the traffics regarding attack and normal are captured on the server.  ...  Then, representative features of traffic are extracted and used by the Multi-Layer Perceptron model of Artificial Neural Network to classify the attack and normal traffic.  ...  Due tothe increasing popularity and importance of Cloud, in this research a method using artificial neural network as classifier is proposed to detect brute force attack on encrypted and tunnelled traffic  ... 
doi:10.3844/jcssp.2014.2029.2036 fatcat:6sjemtrvwbgu3hqwhypme7uaze

Head in the Clouds

John Delaney
2020 IEEE Systems Man and Cybernetics Magazine  
cortex itself, that the network is being attacked," Emondi says.  ...  "They would then wirelessly transmit encoded information to and from a cloud-based supercomputer network for real-time brainstate monitoring and data extraction."  ... 
doi:10.1109/msmc.2020.2978292 fatcat:2uoitog77zgile3q43w7nbsfxm

A Survey of Denial-of-Service and Distributed Denial of Service Attacks and Defenses in Cloud Computing

Adrien Bonguet, Martine Bellaiche
2017 Future Internet  
Cloud Computing is a computing model that allows ubiquitous, convenient and on-demand access to a shared pool of highly configurable resources (e.g., networks, servers, storage, applications and services  ...  In this paper, new types of DoS and DDoS attacks in Cloud Computing are explored, especially the XML-DoS and HTTP-DoS attacks, and some possible detection and mitigation techniques are examined.  ...  [36] explain that firewalls protect the front access points of Clouds and are treated as the first line of defense.  ... 
doi:10.3390/fi9030043 fatcat:zhhl36zts5g7rkstbitjj2kde4

A Review on Cyber Security and the Fifth Generation Cyberattacks

A. Saravanan, S. Sathya Bama
2016 Oriental journal of computer science and technology  
The analysis made for cyberattacks and their statistics shows the intensity of the attacks.  ...  Various cyber security threats are presented along with the machine learning algorithms that can be applied on cyberattacks detection.  ...  Acknowledgements This research has not received any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors declare no conflict of interest.  ... 
doi:10.13005/ojcst12.02.04 fatcat:e4ro2yahenherk5i7j57gl4bcq

New Security State Awareness Model for IoT Devices With Edge Intelligence

Wenxin Lei, Hong Wen, Wenjing Hou, Xinchen Xu
2021 IEEE Access  
It should be explained that the processing time based on the cloud computing paradigm needs to add the network latency from edge to cloud, as shown in the figure, which is approaching 600 ms.  ...  On the other hand, the processing time of the state awareness for event detection under the complex network approach proposed in this paper is higher than that of the SVM approach based on the cloud computing  ... 
doi:10.1109/access.2021.3075220 fatcat:yoykzk5hezbsvovlijnzks66qm

Anomaly Mining – Past, Present and Future [article]

Leman Akoglu
2021 arXiv   pre-print
In this article, I focus on two areas, (1) point-cloud and (2) graph-based anomaly mining.  ...  On the other hand, and in contrast to point-cloud OD, anomaly definitions for GAD vary widely.  ...  Recent advances in deep neural networks (NNs) have also been carried over to this area for deep OD.  ... 
arXiv:2105.10077v2 fatcat:znvvz6ewpbdpjhnhp35kudvzlu

Secure Threat Information Exchange across the Internet of Things for Cyber Defense in a Fog Computing Environment

Mihai-Gabriel IONITA, Victor-Valeriu PATRICIU
2016 Informatică economică  
The agents are based on a neural network which takes actions based on risk assessment inspired by the human immune system.  ...  The fog computing paradigm enhances the use cases of the already used cloud computing systems by bringing all the needed resources to the end-users towards the edge of the network.  ...  Our agents, based on the artificial neural network are especially designed to detect APT attacks.  ... 
doi:10.12948/issn14531305/20.3.2016.02 fatcat:zvofphd5avgohd6iancygyva5a

Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applications

Wenjie Ruan, Xinping Yi, Xiaowei Huang
2021 Proceedings of the 30th ACM International Conference on Information & Knowledge Management  
This tutorial will particularly highlight state-of-the-art techniques in adversarial attacks and robustness verification of deep neural networks (DNNs).  ...  We will also introduce some effective countermeasures to improve robustness of deep learning models, with a particular focus on adversarial training.  ...  In the end of this part, we will also briefly introduce some adversarial attacks on other domains, including attacks on sentiment analysis systems [13] , attacks on 3D point cloud models [9] , attacks  ... 
doi:10.1145/3459637.3482029 fatcat:ekos2t5jmfgahpim76txpf7qxu

An Enhanced OLSR Protocol to Improve Performance of UAV in Wireless Mesh Network

2019 International journal of recent technology and engineering  
In a wireless mesh network, described about the optimized link state routing protocol (OLSR) and works on network parameters.  ...  to complex networks.  ...  ., 2018 [22] addressed a research on inter-cloud interconnection in wireless mesh networks.  ... 
doi:10.35940/ijrte.d7797.118419 fatcat:5dgqliyaqbek5pjiczyse2o4yu

Machine Learning Systems for Intelligent Services in the IoT: A Survey [article]

Wiebke Toussaint, Aaron Yi Ding
2020 arXiv   pre-print
It covers the latest developments (up to 2020) on scaling and distributing ML across cloud, edge, and IoT devices.  ...  This survey moves beyond existing ML algorithms and cloud-driven design to investigate the less-explored systems, scaling and socio-technical aspects for consolidating ML and IoT.  ...  Backdoor attacks on neural network classifiers occur when a pretrained model works well on regular inputs, but provides spurious outputs for specific inputs that are only known by the attacker.  ... 
arXiv:2006.04950v3 fatcat:xrjcioqkrrhpvgmwmutiajgfbe
« Previous Showing results 1 — 15 out of 4,923 results