Zhilin Zhang

Dr. Zhilin Zhang

Amazon (United States), Seattle, Washington, USA
orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 June 2017
Verified
Conference Paper
Closed access
WLWeipeng Lin
Weipeng Lin
KWKe Wang
Ke Wang
Zhilin ZhangZhilin Zhang
Zhilin Zhang
HCHong Chen
Hong Chen
Cloud computing has emerged as a compelling vision for managing data and delivering query answering capability over the internet. This new way of computing also poses a real risk of disclosing confidential information to the cloud. Searchable encryption addresses this issue by allowing the cloud to compute the answer to a query based on the cipher texts of data and queries. Thanks to its inner product preservation property, the asymmetric scalar-product-preserving encryption (ASPE) has been adopted and enhanced in a growing number of works toperform a variety of queries and tasks in the cloud computingsetting. However, the security property of ASPE and its enhancedschemes has not been studied carefully. In this paper, we show acomplete disclosure of ASPE and several previously unknownsecurity risks of its enhanced schemes. Meanwhile, efficientalgorithms are proposed to learn the plaintext of data and queriesencrypted by these schemes with little or no knowledge beyondthe ciphertexts. We demonstrate these risks on real data sets.

26 May 2020 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
XDXiaofeng Ding
Xiaofeng Ding
HFHongbiao Fang
Hongbiao Fang
Zhilin ZhangZhilin Zhang
Zhilin Zhang
KCKim-Kwang Raymond Choo
Kim-Kwang Raymond Choo
HJHai Jin
Hai Jin
Deep learning is increasingly popular, partly due to its widespread application potential, such as in civilian, government and military domains. Given the exacting computational requirements, cloud computing has been utilized to host user data and model. However, such an approach has potential privacy implications. Therefore, in this paper, we propose a method to protect user's privacyin the inference phase of deep learning workflow. Specifically, we use an intermediate layer to separate the entire neural network into two parts, which are respectively deployed on the user device and the cloud server. The encoder, deployed on the user device, is usedfor raw data transformation, which removes the need for users to upload raw data to the cloud directly. However, we also demonstrate there exists potential for privacy leakage in the intermediate features of the neural network through two concrete experiments. In other words, the encoder on its own does not adequately achieve privacy protection. By imposing privacy constraints during the feature extraction, we can reduce thecontribution of the extracted features to the privacy leakage. In this way, privacy protection capability of the encoder can be further strengthened. We then demonstrate the effectiveness of our method using a large number of experiments.

30 January 2019
Verified
Conference Paper
Open access
PDF file
DLDugang Liu
Dugang Liu
CLChen Lin
Chen Lin
Zhilin ZhangZhilin Zhang
Zhilin Zhang
YXYanghua Xiao
Yanghua Xiao
HTHanghang Tong
Hanghang Tong

17 October 2018
Verified
Conference Paper
Open access
PDF file
Zhilin ZhangZhilin Zhang
Zhilin Zhang
KWKe Wang
Ke Wang
CLChen Lin
Chen Lin
WLWeipeng Lin
Weipeng Lin

10 July 2024
Verified
Conference Paper
Open access
YLYang Li
Yang Li
QZQi'AO Zhao
Qi'AO Zhao
CLChen Lin
Chen Lin
JSJinsong Su
Jinsong Su
Zhilin ZhangZhilin Zhang
Zhilin Zhang
Multi-modal Recommendation Systems (MRSs) utilize diverse modalities, such as image and text, to enrich item representations and enhance recommendation accuracy. Current MRSs overlook the large misalignment between multi-modal content features and ID embeddings. While bidirectional alignment between visual and textual modalities has been extensively studied in large multi-modal models, this study suggests that multi-modal alignment in MRSs should be in a one-way direction. A plug-and-play framework is presented, called FEedback-orienTed mulTi-modal aLignmEnt (FETTLE). FETTLE contains three novel solutions: (1) it automatically determines item-level alignment direction between each pair of modalities based on estimated user feedback; (2) it coordinates the alignment directions among multiple modalities; (3) it implements cluster-level alignment from both user and item perspectives for more stable alignments. Extensive experiments on three real datasets demonstrate that FETTLE significantly improves various backbone models. Conventional collaborative filtering models are improved by 24.79%-62.79%, and recent MRSs are improved by 5.91% - 20.11%.

3 November 2019
Verified
Conference Paper
Open access
Zhilin ZhangZhilin Zhang
Zhilin Zhang
KWKe Wang
Ke Wang
WLWeipeng Lin
Weipeng Lin
AFAda Wai-Chee Fu
Ada Wai-Chee Fu
RWRaymond Chi-Wing Wong
Raymond Chi-Wing Wong
We consider the following secure data retrieval problem: a client outsources encrypted data blocks to a semi-trusted cloud server and later retrieves blocks without disclosing access patterns. Existing PIR and ORAM solutions suffer from serious performance bottlenecks in terms of communication or computation costs. To help eliminate this void, we introduce "access pattern unlinkability'' that separates access pattern privacy into short-term privacy at individual query level and long-term privacy at query distribution level. This new security definition provides tunable trade-offs between privacy and query performance. We present an efficient construction, called SBR protocol, using PIR and Oblivious Shuffling to enable secure data retrieval while satisfying access pattern unlinkability. Both analytical and empirical analysis show that SBR exhibits flexibility and usability in practice.