arXiv subjects

Dengpan Ye

Publications and source records attributed to Dengpan Ye.

23 records · Page 2Linked to original sources

arXiv2020

Detection Defense Against Adversarial Attacks with Saliency Map

It is well established that neural networks are vulnerable to adversarial examples, which are almost imperceptible on human vision and can cause the deep models misbehave. Such phenomenon may lead to severely inestimable consequences in the safety and security critical applications. Existing defenses are trend to harden the robustness of models against adversarial attacks, e.g., adversarial training technology. However, these are usually intractable to implement due to the high cost of re-training and the cumbersome operations of altering the model architecture or parameters. In this paper, we discuss the saliency map method from the view of enhancing model interpretability, it is similar to introducing the mechanism of the attention to the model, so as to comprehend the progress of object identification by the deep networks. We then propose a novel method combined with additional noises and utilize the inconsistency strategy to detect adversarial examples. Our experimental results of some representative adversarial attacks on common datasets including ImageNet and popular models show that our method can detect all the attacks with high detection success rate effectively. We compare it with the existing state-of-the-art technique, and the experiments indicate that our method is more general.

cs.LG↗

arXiv2019

Augmenting Encrypted Search: A Decentralized Service Realization with Enforced Execution

Searchable symmetric encryption (SSE) allows the data owner to outsource an encrypted database to a remote server in a private manner while maintaining the ability for selectively search. So far, most existing solutions focus on an honest-but-curious server, while security designs against a malicious server have not drawn enough attention. A few recent works have attempted to construct verifiable SSE that enables the data owner to verify the integrity of search results. Nevertheless, these verification mechanisms are highly dependent on specific SSE schemes, and fail to support complex queries. A general verification mechanism is desired that can be applied to all SSE schemes. In this work, instead of concentrating on a central server, we explore the potential of the smart contract, an emerging blockchain-based decentralized technology, and construct decentralized SSE schemes where the data owner can receive correct search results with assurance without worrying about potential wrongdoings of a malicious server. We study both public and private blockchain environments and propose two designs with a trade-off between security and efficiency. To better support practical applications, the multi-user setting of SSE is further investigated where the data owner allows authenticated users to search keywords in shared documents. We implement prototypes of our two designs and present experiments and evaluations to demonstrate the practicability of our decentralized SSE schemes.

cs.CR↗

arXiv2019

Delay-Distance Correlation Study for IP Geolocation

Although many classical IP geolocation algorithms are suitable to rich-connected networks, their performances are seriously affected in poor-connected networks with weak delay-distance correlation. This paper tries to improve the performances of classical IP geolocation algorithms by finding rich-connected sub-networks inside poor-connected networks. First, a new delay-distance correlation model (RTD-Corr model) is proposed. It builds the relationship between delay-distance correlation and actual network factors such as the tortuosity of the network path, the ratio of propagation delay and the direct geographical distance. Second, based on the RTD-Corr model and actual network characteristics, this paper discusses about how to find rich-connected networks inside China Internet which is a typical actual poor-connected network. Then we find rich-connected sub-networks of China Internet through a large-scale network measurement which covers three major ISPs and thirty provinces. At last, based on the founded rich-connected sub-networks, we modify two classical IP geolocation algorithms and the experiments in China Internet show that their accuracy is significantly increased.

cs.NI↗

arXiv2019

Heard More Than Heard: An Audio Steganography Method Based on GAN

Audio steganography is a collection of techniques for concealing the existence of information by embedding it within a non-secret audio, which is referred to as carrier. Distinct from cryptography, the steganography put emphasis on the hiding of the secret existence. The existing audio steganography methods mainly depend on human handcraft, while we proposed an audio steganography algorithm which automatically generated from adversarial training. The method consists of three neural networks: encoder which embeds the secret message in the carrier, decoder which extracts the message, and discriminator which determine the carriers contain secret messages. All the networks are simultaneously trained to create embedding, extracting and discriminating process. The system is trained with different training settings on two datasets. Competed the majority of audio steganographic schemes, the proposed scheme could produce high fidelity steganographic audio which contains secret audio. Besides, the additional experiments verify the robustness and security of our algorithm.

cs.MM↗

arXiv2017

NEXT: A Neural Network Framework for Next POI Recommendation

The task of next POI recommendation has been studied extensively in recent years. However, developing an unified recommendation framework to incorporate multiple factors associated with both POIs and users remains challenging, because of the heterogeneity nature of these information. Further, effective mechanisms to handle cold-start and endow the system with interpretability are also difficult topics. Inspired by the recent success of neural networks in many areas, in this paper, we present a simple but effective neural network framework for next POI recommendation, named NEXT. NEXT is an unified framework to learn the hidden intent regarding user's next move, by incorporating different factors in an unified manner. Specifically, in NEXT, we incorporate meta-data information and two kinds of temporal contexts (i.e., time interval and visit time). To leverage sequential relations and geographical influence, we propose to adopt DeepWalk, a network representation learning technique, to encode such knowledge. We evaluate the effectiveness of NEXT against state-of-the-art alternatives and neural networks based solutions. Experimental results over three publicly available datasets demonstrate that NEXT significantly outperforms baselines in real-time next POI recommendation. Further experiments demonstrate the superiority of NEXT in handling cold-start. More importantly, we show that NEXT provides meaningful explanation of the dimensions in hidden intent space.

cs.IR↗