Shenghong Li

Shenghong Li

Auto-generated profile, based on publicly available information

Shenghong Li

School of Computer Science, Shanghai Jiao Tong University, Shanghai, China

Research Focus & Metrics

Academic Career Length

24Years (2003-Present)

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

Representative Publications

View All Publications
26 July 2017
Verified
Book Chapter
Closed access
PDF file
XRXudie Ren
Xudie Ren
HGHaonan Guo
Haonan Guo
SLShenghong Li
Shenghong Li
SWShilin Wang
Shilin Wang
JLJianhua Li
Jianhua Li

10 May 2022 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
Gaolei LiGaolei Li
Gaolei Li
Jun WuJun Wu
Jun Wu
SLShenghong Li
Shenghong Li
WYWu Yang
Wu Yang
CLChanglian Li
Changlian Li
Software-defined industrial Internet of things (SD-IIoT) exploits federated learning to process the sensitive data at edges, while adaptive poisoning attacks threat the security of SD-IIoT. To address this problem, this article proposes a multi-tentacle federated learning (MTFL) framework, which is essential to guarantee the trustness of training data in SD-IIoT. In MTFL, participants with similar learning tasks are assigned to the same tentacle group. To identify adaptive poisoning attacks, a tentacle distribution-based efficient poisoning attack detection (TD-EPAD) algorithm is presented. And also, to minimize the impact of adaptive poisoning data, a stochastic tentacle data exchanging (STDE) protocol is also proposed. Simultaneously, to protect the tentacle’s privacy in STDE, all exchanged data will be processed by differential privacy technology. A MTFL prototype system is implemented, which provides extensive ablation experiments and comparison experiments, demonstrating that the accuracy of the global model under attack scenario can be improved with 40%.

31 August 2018 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
CDChong Di
Chong Di
BZBo Zhang
Bo Zhang
QLQilian Liang
Qilian Liang
SLShenghong Li
Shenghong Li
YGYing Guo
Ying Guo
The machine-to-machine (M2M) communications, which achieve the implementation of Internet of Things (IoT), can be carried over wireless cellular networks. The massive random access in M2M communications will cause radio access network congestion in the base station (BS), leading to sharp deterioration in access delay and access probability. Access class barring (ACB) that can directly control the flow of machine-type communication (MTC) devices by an ACB factor is an efficient scheme to prevent the BS from traffic overload. In wireless cellular networks, the random access resources (i.e., preambles) are shared by M2M and human-to-human (H2H) devices, and research on ACB scheme ordinarily assumes that a restricted number of preambles are assigned to M2M traffic. However, when suffering from massive access in M2M communications, it is desirable to rapidly satisfy the access requests from MTC devices using all available preambles, especially in time-sensitive IoT scenarios. In this paper, we study the massive access problem in M2M traffic centered scenarios where M2M and H2H traffic can apply for all available preambles without distinction. Utilizing the self-adaptive learning property of learning automata, we further propose a novel learning automata based ACB (LA-ACB) scheme. Simulation results show that the LA-ACB scheme achieves the performance close to theoretical optimality. The BS equipped with the LA-ACB scheme can effectively control the M2M traffic by dynamically adjusting the ACB factor under the interference of H2H traffic and provide quality services for both M2M and H2H traffic.

29 September 2021 in IEEE Transactions on Information Forensics and Security
Verified
Journal Article
Closed access
ZYZhicong Yan
Zhicong Yan
Jun WuJun Wu
Jun Wu
Gaolei LiGaolei Li
Gaolei Li
SLShenghong Li
Shenghong Li
Mohsen GuizaniMohsen Guizani
Mohsen Guizani
Semi-Supervised Learning (SSL) is a powerful derivative for humans to discover the hidden knowledge, and will be a great substitute for data taggers. Although the availability of unlabeled data rises up a huge passion to SSL, the untrustness of unlabeled data leads to many unknown security risks. In this paper, we first identify an insidious backdoor threat of SSL where unlabeled training data are poisoned by backdoor methods migrated from supervised settings. Then, to further exploit this threat, a Deep Neural Backdoor (DeNeB) scheme is proposed, which requires less data poisoning budgets and produces stronger backdoor effectiveness. By poisoning a fraction of our unlabeled training data, the DeNeB achieves the illegal manipulation on the trained model without modifying the training process. Finally, an efficient detection-and-purification defense (DePuD) framework is proposed to thwart the proposed scheme. In DePuD, we construct a deep detector to locate trigger patterns in the unlabeled training data, and perform secured SSL training with purified unlabeled data where the detected trigger patterns are obfuscated. Extensive experiments based on benchmark datasets are performed to demonstrate the huge threatening of DeNeB and the effectiveness of DePuD. To the best of our knowledge, this is the first work to achieve the backdoor and its defense in semi-supervised learning.

6 June 2017 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
HGHaonan Guo
Haonan Guo
SLShenghong Li
Shenghong Li
BLBin Li
Bin Li
YMYinghua Ma
Yinghua Ma
XRXudie Ren
Xudie Ren
Deep neural network are one of the most powerful model for machine learning, which can learn the underlying patterns automatically from a large amount of data. So it can be extensively used in more and more Internet-of-Things (IoT) applications. However, the training of deep models is difficult, suffering from overfitting and gradient vanishing problem. Besides, the large amount of parameters and multiplication operations make it impractical for most deep learning models to directly execute on target hardware. In this paper, we propose a method of gradually pruning the weakly connected weights to improve the traditional stochastic gradient descent. And we adopt a reinforcement learning method called learning automata to find the weakly connected weights on account of its strong policy-making ability in stochastic and nonstationary environment. Our proposed method can learn a more effective and sparsely connected architecture during training from the initially fully connected neural networks. The experiments on MNIST show that our method have stronger power to defeat overfitting and can get better generalization performance on test set. Meanwhile, the thin and sparsely connected model we get can be more suitable for IoT applications.

17 November 2022 in IEEE Transactions on Dependable and Secure Computing
Verified
Journal Article
Closed access
GRGe Ren
Ge Ren
Jun WuJun Wu
Jun Wu
Gaolei LiGaolei Li
Gaolei Li
SLShenghong Li
Shenghong Li
Mohsen GuizaniMohsen Guizani
Mohsen Guizani
Artificial intelligence (AI)-based cybersecurity services offer significant promise in many scenarios, including malware detection, content supervision, and so on. Meanwhile, many commercial and government applications have raised the need for intellectual property protection of using deep neural network (DNN). Existing studies (e.g., watermarking techniques) on intellectual property protection only aim at inserting secret information into DNNs, allowing producers to detect whether the given DNN infringes on their own copyrights. However, since the availability protection of learning models is rarely considered, the piracy model can still work with high accuracy. In this paper, a novel model locking (M-LOCK) scheme for the DNN is proposed to enhance its availability protection, where the DNN produces poor accuracy if a specific token is absent, while it maps only the tokenized inputs into correct predictions. The proposed scheme performs the verification process during the DNN inference operation, actively protecting models' intellectual property copyright at each query. Specifically, to train the token-sensitive decision-making boundaries of DNNs, a data poisoning-based model manipulation (DPMM) method is also proposed, which minimizes the correlation between the dummy outputs and correct predictions. Extensive experiments demonstrate the proposed scheme could achieve high reliability and effectiveness across various benchmark datasets as well as typical model protection methods.