The-Anh Ta

The-Anh Ta

Auto-generated profile, based on publicly available information

The-Anh Ta

CSIRO’s Data61

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
22 September 2022
Verified
Book Chapter
Closed access
PDF file
TNTuong Ngoc Nguyen
Tuong Ngoc Nguyen
ATAnh The Ta
Anh The Ta
HLHuy Quoc Le
Huy Quoc Le
DDDung Hoang Duong
Dung Hoang Duong
Willy SusiloWilly Susilo
Willy Susilo
SKShinsaku Kiyomoto
Shinsaku Kiyomoto

18 December 2023 in Future Internet
Verified
Journal Article
Open access
PDF file
DNDuy Tung Khanh Nguyen
Duy Tung Khanh Nguyen
DDDung Hoang Duong
Dung Hoang Duong
Willy SusiloWilly Susilo
Willy Susilo
Yang-Wai ChowYang-Wai Chow
Yang-Wai Chow
TTThe Anh Ta
The Anh Ta
Homomorphic encryption (HE) has emerged as a pivotal technology for secure neural network inference (SNNI), offering privacy-preserving computations on encrypted data. Despite active developments in this field, HE-based SNNI frameworks are impeded by three inherent limitations. Firstly, they cannot evaluate non-linear functions such as ReLU, the most widely adopted activation function in neural networks. Secondly, the permitted number of homomorphic operations on ciphertexts is bounded, consequently limiting the depth of neural networks that can be evaluated. Thirdly, the computational overhead associated with HE is prohibitively high, particularly for deep neural networks. In this paper, we introduce a novel paradigm designed to address the three limitations of HE-based SNNI. Our approach is an interactive approach that is solely based on HE, called iLHE. Utilizing the idea of iLHE, we present two protocols: ReLU, which facilitates the direct evaluation of the ReLU function on encrypted data, tackling the first limitation, and HeRefresh, which extends the feasible depth of neural network computations and mitigates the computational overhead, thereby addressing the second and third limitations. Based on HeReLU and HeRefresh protocols, we build a new framework for SNNI, named HeFUN. We prove that our protocols and the HeFUN framework are secure in the semi-honest security model. Empirical evaluations demonstrate that HeFUN surpasses current HE-based SNNI frameworks in multiple aspects, including security, accuracy, the number of communication rounds, and inference latency. Specifically, for a convolutional neural network with four layers on the MNIST dataset, HeFUN achieves 99.16% accuracy with an inference latency of 1.501 s, surpassing the popular HE-based framework CryptoNets proposed by Gilad-Bachrach, which achieves 98.52% accuracy with an inference latency of 3.479 s.

5 January 2024
Verified
Book Chapter
Closed access
PDF file
XKXuan Thanh Khuc
Xuan Thanh Khuc
ATAnh The Ta
Anh The Ta
Willy SusiloWilly Susilo
Willy Susilo
DDDung Hoang Duong
Dung Hoang Duong
FGFuchun Guo
Fuchun Guo
SKShinsaku Kiyomoto
Shinsaku Kiyomoto

21 November 2024
Verified
Book Chapter
Closed access
PDF file
HPHoang Pham
Hoang Pham
TTThe-Anh Ta
The-Anh Ta
ATAnh Tran
Anh Tran
KDKhoa D. Doan
Khoa D. Doan

13 October 2025
Verified
Conference Paper
Open access
TTThe-Anh Ta
The-Anh Ta
SCSid Chau
Sid Chau

16 July 2024 in arXiv
Verified
Preprint
Open access
PDF file
QNQuang H. Nguyen
Quang H. Nguyen
NNNguyen Ngoc-Hieu
Nguyen Ngoc-Hieu
TTThe-Anh Ta
The-Anh Ta
TNThanh Nguyen-Tang
Thanh Nguyen-Tang
KWKOK-SENG WONG
KOK-SENG WONG
KDKhoa D. Doan
Khoa D. Doan
Deep neural networks are vulnerable to backdoor attacks, a type of adversarial attack that poisons the training data to manipulate the behavior of models trained on such data. Clean-label attacks are a more stealthy form of backdoor attacks that can perform the attack without changing the labels of poisoned data. Early works on clean-label attacks added triggers to a random subset of the training set, ignoring the fact that samples contribute unequally to the attack's success. This results in high poisoning rates and low attack success rates. To alleviate the problem, several supervised learning-based sample selection strategies have been proposed. However, these methods assume access to the entire labeled training set and require training, which is expensive and may not always be practical. This work studies a new and more practical (but also more challenging) threat model where the attacker only provides data for the target class (e.g., in face recognition systems) and has no knowledge of the victim model or any other classes in the training set. We study different strategies for selectively poisoning a small set of training samples in the target class to boost the attack success rate in this setting. Our threat model poses a serious threat in training machine learning models with third-party datasets, since the attack can be performed effectively with limited information. Experiments on benchmark datasets illustrate the effectiveness of our strategies in improving clean-label backdoor attacks.