Search

Halide perovskite materials have broad prospects for applications in various fields such as solar cells, LED devices, photodetectors, fluorescence labeling, bioimaging, and photocatalysis due to their bandgap characteristics. This study compiled experimental data from the published literature and utilized the excellent predictive capabilities, low overfitting risk, and strong robustness of ensemble learning models to analyze the bandgaps of halide perovskite compounds. The results demonstrate the effectiveness of ensemble learning decision tree models, especially the gradient boosting decision tree model, with a root mean square error of 0.090 eV, a mean absolute error of 0.053 eV, and a determination coefficient of 93.11%. Research on data related to ratios calculated through element molar quantity normalization indicates significant influences of ions at the X and B positions on the bandgap. Additionally, doping with iodine atoms can effectively reduce the intrinsic bandgap, while hybridization of the s and p orbitals of tin atoms can also decrease the bandgap. The accuracy of the model is validated by predicting the bandgap of the photovoltaic material MASn1−xPbxI3. In conclusion, this study emphasizes the positive impact of machine learning on material development, especially in predicting the bandgaps of halide perovskite compounds, where ensemble learning methods demonstrate significant advantages. [ABSTRACT FROM AUTHOR]

This paper proposes a system of 6G Internet of Things (IoT) based on downlink non-orthogonal multiple access (NOMA) technology, where the base station (BS) allows signals of the same frequency to serve users at different distances. In particular, we study a cooperative MIMO-NOMA system based on downlink simultaneous wireless information and power transfer (SWIPT) assistance. To improve the overall performance, we employ machine learning to optimize user-pairing and radio resource allocation. At the end of the paper, the simulation results are obtained, which fully prove that the MIMO-NOMA system constructed in this paper is correct in theory and can be realized in practice. [ABSTRACT FROM AUTHOR]

The performance of deep learning is heavily influenced by the size of the learning samples, whose labeling process is time consuming and laborious. Deep learning algorithms typically assume that the training and prediction data are independent and uniformly distributed, which is rarely the case given the attributes and properties of different data sources. In remote sensing images, representations of urban land surfaces can vary across regions and by season, demanding rapid generalization of these surfaces in remote sensing data. In this study, we propose Meta-FSEO, a novel model for improving the performance of few-shot remote sensing scene classification in varying urban scenes. The proposed Meta-FSEO model deploys self-supervised embedding optimization for adaptive generalization in new tasks such as classifying features in new urban regions that have never been encountered during the training phase, thus balancing the requirements for feature classification tasks between multiple images collected at different times and places. We also created a loss function by weighting the contrast losses and cross-entropy losses. The proposed Meta-FSEO demonstrates a great generalization capability in remote sensing scene classification among different cities. In a five-way one-shot classification experiment with the Sentinel-1/2 Multi-Spectral (SEN12MS) dataset, the accuracy reached 63.08%. In a five-way five-shot experiment on the same dataset, the accuracy reached 74.29%. These results indicated that the proposed Meta-FSEO model outperformed both the transfer learning-based algorithm and two popular meta-learning-based methods, i.e., MAML and Meta-SGD. [ABSTRACT FROM AUTHOR]

Membership inference attacks (MIA) exploit the fact that deep learning algorithms leak information about their training data through the learned model. It has been treated as an indicator which reveals the privacy leakage of machine learning models. In this work, we aim to understand the advantage achieved by White-box MIA, and defend against White-box MIA. Firstly, we estimate the KL divergence on the hidden layers' features between training set and test set as the internal generalization gap. By comparing the internal generalization gap and the generalization gap on the output layer, we raise two insights including (1) the existence of larger generalization gaps on hidden layers and (2) feasibility of generalization gap minimization in defending White-box MIA. Based on our insights, we further design a novel defense method named Nirvana. It intentionally minimizes generalization gap to defend White-box MIA. Formally, Nirvana works by selecting a hidden layer with large generalization gaps and executing a multi-samples convex combination among features on the layer during the training to defend against White-box MIA. Finally, we empirically evaluate Nirvana with state-of-the-art defense methods on CIFAR100 dataset, Purchase100 dataset, and Texas dataset. The experiment results show that Nirvana achieves a trade-off between utility and privacy. It can defend both White-box MIA and Black-box MIA while the test accuracy of the model is maintained. It outperforms previous defense methods in defending against White-box MIA. [ABSTRACT FROM AUTHOR]