Sub-diffraction terahertz backpropagation compressive imaging

Terahertz single-pixel imaging (TSPI) has garnered significant attention due to its simplicity and cost-effectiveness. However, the relatively long wavelength of THz waves limits sub-diffraction-scale imaging resolution. Although TSPI technique can achieve sub-wavelength resolution, it requires harsh experimental conditions and time-consuming processes. Here, we propose a sub-diffraction THz backpropagation compressive imaging technique. We illuminate the object with monochromatic continuous-wave THz radiation. The transmitted THz wave is modulated by prearranged patterns generated on the back surface of a 500-{\mu}m-thick silicon wafer, realized through photoexcited carriers using a 532-nm laser. The modulated THz wave is then recorded by a single-element detector. An untrained neural network is employed to iteratively reconstruct the object image with an ultralow compression ratio of 1.5625% under a physical model constraint, thus reducing the long sampling times. To further suppress the diffraction-field effects, embedded with the angular spectrum propagation (ASP) theory to model the diffraction of THz waves during propagation, the network retrieves near-field information from the object, enabling sub-diffraction imaging with a spatial resolution of ~{\lambda}0/7 ({\lambda}0 = 833.3 {\mu}m at 0.36 THz) and eliminating the need for ultrathin photomodulators. This approach provides an efficient solution for advancing THz microscopic imaging and addressing other inverse imaging challenges.

Lurking in the shadows: Unveiling Stealthy Backdoor Attacks against Personalized Federated Learning

Federated Learning (FL) is a collaborative machine learning technique where multiple clients work together with a central server to train a global model without sharing their private data. However, the distribution shift across non-IID datasets of clients poses a challenge to this one-model-fits-all method hindering the ability of the global model to effectively adapt to each client's unique local data. To echo this challenge, personalized FL (PFL) is designed to allow each client to create personalized local models tailored to their private data. While extensive research has scrutinized backdoor risks in FL, it has remained underexplored in PFL applications. In this study, we delve deep into the vulnerabilities of PFL to backdoor attacks. Our analysis showcases a tale of two cities. On the one hand, the personalization process in PFL can dilute the backdoor poisoning effects injected into the personalized local models. Furthermore, PFL systems can also deploy both server-end and client-end defense mechanisms to strengthen the barrier against backdoor attacks. On the other hand, our study shows that PFL fortified with these defense methods may offer a false sense of security. We propose \textit{PFedBA}, a stealthy and effective backdoor attack strategy applicable to PFL systems. \textit{PFedBA} ingeniously aligns the backdoor learning task with the main learning task of PFL by optimizing the trigger generation process. Our comprehensive experiments demonstrate the effectiveness of \textit{PFedBA} in seamlessly embedding triggers into personalized local models. \textit{PFedBA} yields outstanding attack performance across 10 state-of-the-art PFL algorithms, defeating the existing 6 defense mechanisms. Our study sheds light on the subtle yet potent backdoor threats to PFL systems, urging the community to bolster defenses against emerging backdoor challenges.