Search

Gradient inversion attacks (GIAs) have posed significant challenges to the emerging paradigm of distributed learning, which aims to reconstruct the private training data of clients (participating parties in distributed training) through the shared parameters. For counteracting GIAs, a large number of privacy-preserving methods for distributed learning scenario have emerged. However, these methods have significant limitations, either compromising the usability of global model or consuming substantial additional computational resources. Furthermore, despite the extensive efforts dedicated to defense methods, the underlying causes of data leakage in distributed learning still have not been thoroughly investigated. Therefore, this paper tries to reveal the potential reasons behind the successful implementation of existing GIAs, explore variations in the robustness of models against GIAs during the training process, and investigate the impact of different model structures on attack performance. After these explorations and analyses, this paper propose a plug-and-play GIAs defense method, which augments the training data by a designed vicinal distribution. Sufficient empirical experiments demonstrate that this easy-to-implement method can ensure the basic level of privacy without compromising the usability of global model.

Model inversion attacks aim to reveal information about sensitive training data of AI models, which may lead to serious privacy leakage. However, existing attack methods have limitations in reconstructing training data with higher feature fidelity. In this article, we propose an evolutionary model inversion attack approach (EvoMI) and empirically demonstrate that combined with the systematic search in the multi-degree-of-freedom latent space of the generative model, the simple use of an evolutionary algorithm can effectively improve the attack performance. Concretely, at first, we search for latent vectors which can generate images close to the attack target in the latent space with low-degree of freedom. Generally, the low-freedom constraint will reduce the probability of getting a local optima compared to existing methods that directly search for latent vectors in the high-freedom space. Consequently, we introduce a mutation operation to expand the search domain, thus further reduce the possibility of obtaining a local optima. Finally, we treat the searched latent vectors as the initial values of the post-processing and relax the constraint to further optimize the latent vectors in a higher-freedom space. Our proposed method is conceptually simple and easy to implement, yet it achieves substantial improvements and outperforms the state-of-the-art methods significantly.

Privacy-preserving machine learning requires that models do not reveal any private information about their training data. However, model inversion attacks (MIAs), which aim to recover the features of training data, pose a huge threat to the security of AI models. Most existing MIAs assume that the target model is white-box, but most models deployed in reality are black-box, and these models can only be accessed like an oracle. There are a few studies for black-box scenarios, but their performance is limited. In this article, we first formulate the MIA problem completely in Bayesian perspective. Second, we propose a novel two-stage MIA approach, the Coarse-to-Fine Model Inversion Attack (C2FMI), which efficiently addresses the MIA problem in the black-box scenario. In stage I of C2FMI, we design a reverse network that constrains the recovered images (also named attacked images) to fall near the manifold space of the training data. In stage II, we design a black-box oriented strategy which further facilitates the attacked images to approach the training data. Empirically, C2FMI achieves a performance that even surpasses existing white-box attack methods. Furthermore, we design the stability analysis method for analyzing the stability of C2FMI along with existing MIAs. Finally, we explore the potential countermeasures which could defend against our attacks.

Caricature is a type of artistic style of human faces that attracts considerable attention in the entertainment industry. So far a few 3D caricature generation methods exist and all of them require some caricature information (e.g., a caricature sketch or 2D caricature) as input. This kind of input, however, is difficult to provide by non-professional users. In this paper, we propose an end-to-end deep neural network model that generates high-quality 3D caricatures directly from a normal 2D face photo. The most challenging issue for our system is that the source domain of face photos (characterized by normal 2D faces) is significantly different from the target domain of 3D caricatures (characterized by 3D exaggerated face shapes and textures). To address this challenge, we: (1) build a large dataset of 5,343 3D caricature meshes and use it to establish a PCA model in the 3D caricature shape space; (2) reconstruct a normal full 3D head from the input face photo and use its PCA representation in the 3D caricature shape space to establish correspondences between the input photo and 3D caricature shape; and (3) propose a novel character loss and a novel caricature loss based on previous psychological studies on caricatures. Experiments including a novel two-level user study show that our system can generate high-quality 3D caricatures directly from normal face photos.

The liquid-solid transitions of (Co2Si+CoSi) and (CoSi+CoSi2) eutectic alloys were realized in drop tube and the rapid eutectic growth mechanism of intermetallic compounds was examined. The experimental and calculated results indicate that with increasing Co content, the intermetallic compound prefers nucleating primarily. The eutectic microstructures experience the transitions of lamellar-anomalous-divorced eutectic with undercooling. In undercooled state, the growth of CoSi intermetallic compound always lags behind others, and no matter how large the undercooling is, this intermetallic compound grows under the solutal diffusion control. The calculated coupled zone demonstrates that (Co2Si+CoSi) eutectic can form within certain undercooling regime, when the composition is in the range from 23.6% to 25.4% Si. And the calculated coupled zone of (CoSi+CoSi2) covers a composition range from 40.8% to 43.8% Si. [ABSTRACT FROM AUTHOR]

Integrating a bioelectrochemical system with CO2electroreduction (CO2ER) can achieve recovery of resources and conversion of value-added chemicals, but it still faces challenge of high overpotential and poor selectivity. Herein, we report a CO2ER catalyst with iron bonded to nitrogen atoms (Fe-Nx) anchored hierarchical porous carbon (Fe SA-NC) by a molecular-confined pyrolysis strategy. Owing to the high surface area and atomic-level Fe-N4sites, Fe SA-NC possessed superior CO2-to-CO conversion performance with a low overpotential of 90 mV, a small Tafel slope of 92 mV dec−1, and high Faradaic efficiency of 95.9% at −0.5 V, superior to almost all previously reported Fe-Nxbased carbon materials for CO2ER. Experimental results manifested the atomic-level Fe-N4sites in carbon frameworks with a single Fe atom coordinating four N atoms. Theoretical calculations revealed Fe-N4sites weaken the free energy for the formation of *COOH intermediate, and the short Fe-C bond length in the structure of *COOH absorbed on Fe-N4sites accelerated the electron transfer from Fe-N4centers to *COOH, thus boosting the reaction kinetics. An integrated device with cathodic Fe SA-NC and bioanode can recover energy and carbon resource from wastewater, delivering maximum current and CO production rate of 1.54 ± 0.05 mA and 33.66 ± 0.58 mmol g−1cath−1.