Zhipeng Zhou

Zhipeng Zhou

Auto-generated profile, based on publicly available information

Zhipeng Zhou

University of Science and Technology of China

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
10 January 2024 in ACM Transactions on Sensor Networks
Verified
Journal Article
Open access
ZZZhipeng Zhou
Zhipeng Zhou
FWFeng Wang
Feng Wang
wei gongwei gong
wei gong
Recently, using deep learning to achieve WiFi-based human activity recognition (HAR) has drawn significant attention. While capable of achieving accurate identification in a single domain (i.e., training and testing in the same consistent WiFi environment), it would become extremely tough when WiFi environments change significantly. As such, domain adversarial neural networks-based approaches have been proposed to handle such diversities across domains, yet often found to share the same limitation in practice: the imbalance between high-capacity of feature extractors and data insufficiency of source domains. This article proposes i-Sample, an intermediate sample generation-based framework, striving to tackle this issue for WiFi-based HAR. i-Sample is mainly designed as two-stage training, where four data augmentation operations are proposed to train a coarse domain-invariant feature extractor in the first stage. In the second stage, we leverage the gradients of classification error to generate intermediate samples to refine the classifiers together with original samples, making i-Sample also capable to be integrated into most domain adversarial adaptation methods without neural network modification. We have implemented a prototype system to evaluate i-Sample, which shows that i-Sample can effectively augment the performance of nowadays mainstream domain adversarial adaptation models for WiFi-based HAR, especially when source domain data is insufficient.

28 June 2024 in IEEE Transactions on Mobile Computing
Verified
Journal Article
Closed access
YPYuhan Pan
Yuhan Pan
ZZZhipeng Zhou
Zhipeng Zhou
wei gongwei gong
wei gong
YFYuguang Fang
Yuguang Fang
Recently, the continuous evolution of deep learning has opened up promising avenues to groundbreaking advancements in wireless sensing systems, which significantly enhance the practical applications of WiFi-based Human Activity Recognition (HAR) systems. However, despite these strides, such systems remain susceptible to adversarial attacks. This article unveils the vulnerability of existing WiFi-based HAR systems to common adversaries, revealing their insufficient robustness. While the intuitive approach is to employ adversarial training to fortify the models, our investigation exposes inherent deficiencies in the current approach. Specifically, we confirm that the strength of perturbations directly influences training outcomes. Moreover, even when confined within a specified perturbation radius, the perturbation strength exhibits variability within a prescribed range, potentially giving rise to “extreme” samples that could compromise training results. To address this challenge, we propose a two-stage Selective Adversarial Training (SAT) approach that integrates model confidence calibration and sample selection. Specifically, we start with calibrating the model and then selectively choose samples from all adversarial examples based on the calibrated confidence outputs that align with the desired criteria for adversarial training. This sample-wise perturbation intensity control effectively prevents the inclusion of inappropriate samples in training, a capability lacking in previous domain-wise perturbation control. Our experiments demonstrate that the proposed fine-grained training method, SAT, is both straightforward and effective in augmenting adversarial training results.

10 January 2024 in IEEE Transactions on Evolutionary Computation
Verified
Journal Article
Closed access
JFJian Feng
Jian Feng
YHYajie He
Yajie He
YPYuhan Pan
Yuhan Pan
ZZZhipeng Zhou
Zhipeng Zhou
SCSi Chen
Si Chen
wei gongwei gong
wei gong
AI-aided Financial Regulation (AIFR) is a practical and significant task, but current solutions have yet to be optimized with customized model designs. Given the privacy concerns surrounding financial data, we aim to employ Neural Architecture Search (NAS) to help non-expert end-users automatically design architectures. The genetic algorithm-based NAS stands out due to its relatively low hardware requirements and robust theoretical foundation. However, constrained by limited data, the model would undergo architecture search on a general regulatory dataset while being deployed on private one owned by each organization. The data distribution of the private dataset may vary from that of public datasets, giving rise to the challenge of data domain shift. To alleviate this problem, we propose a novel fitness evaluation method. When scoring the fitness, we take into account both the architecture’s validation accuracy and its potential for generalization by the metric of loss landscape. In addition, we improve the training paradigm for evaluation, utilizing a prototype-based training paradigm based on embedding distances for classification, allowing for rapid domain adaptation and improve performance on the distribution-shift data. We further introduce GA-TextCNN, a GA-based NAS framework specifically designed for text recognition, enhancing its suitability for text data within AIFR tasks. To demonstrate the effectiveness of our approach, we collect two related datasets and evaluate our method on it. The extensive experiments demonstrate that our method significantly improves baseline models and is effective in solving the AIFR problem.

2 October 2018 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
RZRui Zhang
Rui Zhang
DYDong Yin
Dong Yin
JDJinwen Ding
Jinwen Ding
YLYuhao Luo
Yuhao Luo
WLWei Liu
Wei Liu
ZZZhipeng Zhou
Zhipeng Zhou

24 September 2025 in ACM Transactions on Sensor Networks
Verified
Journal Article
Open access
YWYouquan Wang
Youquan Wang
ZZZhipeng Zhou
Zhipeng Zhou
SWShuai Wang
Shuai Wang
Xianjun DengXianjun Deng
Xianjun Deng
WXWei Xi
Wei Xi
wei gongwei gong
wei gong
WiFi-based human activity recognition (WiFi-based HAR) has emerged as a technology in recent decades, offering convenient and privacy-friendly applications. However, existing frameworks designed for stable environments encounter challenges when faced with changing circumstances and imbalanced training datasets in realistic scenarios. In this article, we address both issues from a unified perspective by exploring a more generalized local minima. Initially, we revisit existing solutions and empirically observe the presence of sharp minima in trained long-tailed WiFi-based HAR models. Consequently, we propose a novel method called Class Region Flattening ( CRF ) to identify class-conditional flat minima. This approach effectively mitigates bias caused by the long-tailed distribution and enhances generalization capabilities in the face of changing circumstances. Furthermore, we introduce a selective flattening operation to prevent optimization conflicts among different activity categories and reduce computational overhead. We integrate CRF into mainstream WiFi-based HAR models and evaluate their performance using our collected WiFi-based HAR dataset. Through extensive experiments, we demonstrate that the incorporation of CRF leads to significant improvements in performance. These findings underscore the effectiveness of CRF in addressing the challenges posed by changing circumstances and imbalanced training datasets in WiFi-based HAR.

11 November 2025 in Pattern Recognition
Verified
Journal Article
Closed access
MYMianzimei Yang
Mianzimei Yang
ZZZhipeng Zhou
Zhipeng Zhou
WGWei Gong
Wei Gong