Frontiers of Computer Science最新文献

Comprehensive comparison results on the above two datasets indicate that the detection improvements proposed in CWYOLO framework for low-light conditions are effective and can stand out among the existing excellent method. In future work, we would explore a more efficient and lightweight network architecture with group convolution to advance the mobile deployment of the detection framework.

Traditional image-sentence cross-modal retrieval methods usually aim to learn consistent representations of heterogeneous modalities, thereby to search similar instances in one modality according to the query from another modality in result. The basic assumption behind these methods is that parallel multi-modal data (i.e., different modalities of the same example are aligned) can be obtained in prior. In other words, the image-sentence cross-modal retrieval task is a supervised task with the alignments as ground-truths. However, in many real-world applications, it is difficult to realign a large amount of parallel data for new scenarios due to the substantial labor costs, leading the non-parallel multi-modal data and existing methods cannot be used directly. On the other hand, there actually exists auxiliary parallel multi-modal data with similar semantics, which can assist the non-parallel data to learn the consistent representations. Therefore, in this paper, we aim at “Alignment Efficient Image-Sentence Retrieval” (AEIR), which recurs to the auxiliary parallel image-sentence data as the source domain data, and takes the non-parallel data as the target domain data. Unlike single-modal transfer learning, AEIR learns consistent image-sentence cross-modal representations of target domain by transferring the alignments of existing parallel data. Specifically, AEIR learns the image-sentence consistent representations in source domain with parallel data, while transferring the alignment knowledge across domains by jointly optimizing a novel designed cross-domain cross-modal metric learning based constraint with intra-modal domain adversarial loss. Consequently, we can effectively learn the consistent representations for target domain considering both the structure and semantic transfer. Furthermore, extensive experiments on different transfer scenarios validate that AEIR can achieve better retrieval results comparing with the baselines.

Federated learning is a promising learning paradigm that allows collaborative training of models across multiple data owners without sharing their raw datasets. To enhance privacy in federated learning, multi-party computation can be leveraged for secure communication and computation during model training. This survey provides a comprehensive review on how to integrate mainstream multi-party computation techniques into diverse federated learning setups for guaranteed privacy, as well as the corresponding optimization techniques to improve model accuracy and training efficiency. We also pinpoint future directions to deploy federated learning to a wider range of applications.

In this study, we have developed a neural network aimed at enhancing the precision of neural distinguishers, demonstrating its capability to surpass DDT-based distinguishers in certain rounds. To extend the scope of our key recovery attack to additional rounds, we have diligently focused on improving both classical differentials and neural distinguishers. Consequently, we have successfully executed practical key recovery attacks on SIMECK32/64, effectively advancing the practical attack threshold by two additional rounds, allowing us to reach up to 17 rounds.

In this paper, we propose the community search framework searching polarized communities via adaptively fusing structure and attribute in attributed signed networks, which searches for two polarized subgraphs on an attributed signed network for given query nodes. We first conduct a analysis by the similarity of attributes between nodes. And we adaptively integrate topology and node attributes into an augmented signed network. Then, a spectral method based on generalized Rayleigh quotient is proposed. Finally, a linear programming problem is designed to detect polarized communities by local eigenspace. Experiments on real-world datasets demonstrate the effectiveness of our method.

Face attribute classification (FAC) is a high-profile problem in biometric verification and face retrieval. Although recent research has been devoted to extracting more delicate image attribute features and exploiting the inter-attribute correlations, significant challenges still remain. Wavelet scattering transform (WST) is a promising non-learned feature extractor. It has been shown to yield more discriminative representations and outperforms the learned representations in certain tasks. Applied to the image classification task, WST can enhance subtle image texture information and create local deformation stability. This paper designs a scattering-based hybrid block, to incorporate frequency-domain (WST) and image-domain features in a channel attention manner (Squeeze-and-Excitation, SE), termed WS-SE block. Compared with CNN, WS-SE achieves a more efficient FAC performance and compensates for the model sensitivity of the small-scale affine transform. In addition, to further exploit the relationships among the attribute labels, we propose a learning strategy from a causal view. The cause attributes defined using the causality-related information can be utilized to infer the effect attributes with a high confidence level. Ablative analysis experiments demonstrate the effectiveness of our model, and our hybrid model obtains state-of-the-art results in two public datasets.

Due to the advantages of high volume of transactions and low resource consumption, Directed Acyclic Graph (DAG)-based Distributed Ledger Technology (DLT) has been considered a possible next-generation alternative to block-chain. However, the security of the DAG-based system has yet to be comprehensively understood. Aiming at verifying and evaluating the security of DAG-based DLT, we develop a Multi-Agent based IOTA Simulation platform called MAIOTASim. In MAIOTASim, we model honest and malicious nodes and simulate the configurable network environment, including network topology and delay. The double-spending attack is a particular security issue related to DLT. We perform the security verification of the consensus algorithms under multiple double-spending attack strategies. Our simulations show that the consensus algorithms can resist the parasite chain attack and partially resist the splitting attack, but they are ineffective under the large weight attack. We take the cumulative weight difference of transactions as the evaluation criterion and analyze the effect of different consensus algorithms with parameters under each attack strategy. Besides, MAIOTASim enables users to perform large-scale simulations with multiple nodes and tens of thousands of transactions more efficiently than state-of-the-art ones.