Federated learning (FL) is a decentralized machine learning framework that prioritizes privacy by allowing clients to train statistical models without sharing their private data, thus eliminating the impact of data fo...
Federated learning (FL) is a decentralized machine learning framework that prioritizes privacy by allowing clients to train statistical models without sharing their private data, thus eliminating the impact of data fortresses. However, the presence of Byzantine attacks, such as data poisoning and backdoor attack, threatens the robustness of FL schemes. Currently, existing mainstream defense methods are susceptible to multiple adaptive attacks, some of which even violate the privacy principle of FL. Furthermore, these defense schemes become less robust when subjected to targeted poisoning attacks with highly non-IID data distributions. In this work, we propose FedNAT, a novel Byzantine-robust FL framework for whittling away these limitations mentioned above. Specifically, FedNAT first performs a privacy-respecting attention refinement on the activation layer outputs of the local uploads. Then, the server scores the local attentions by calculating their Wasserstein distances and clusters them through the k-median algorithm for global attention aggregation, thus rejecting poisoned local attentions for untargeted attacks. After this process, the global attention is transferred to local attention through the FedNAT loss function, which erases backdoors through the distillation concept. We conduct a comprehensive experimental evaluation to demonstrate that FedNAT significantly outperforms existing robust FL schemes in defending against Byzantine poisoning attacks under both IID and highly non-IID data proportions.
Image deblurring task is an ill-posed one, where exists infinite feasible solutions for blurry image. Modern deep learning approaches usually discard the learning of blur kernels and directly employ end-to-end supervi...
详细信息
Maximum likelihood estimation has been widely adopted along with the encoder-decoder framework for video captioning. However, it ignores the structure of sentences and restrains the diversity and distinction of genera...
详细信息
ISBN:
(纸本)9781665487900
Maximum likelihood estimation has been widely adopted along with the encoder-decoder framework for video captioning. However, it ignores the structure of sentences and restrains the diversity and distinction of generated captions. To address this issue, we propose a hard contrastive learning (HCL) method for video captioning. Specifically, built on the encoder-decoder framework, we introduce mismatched pairs to learn a reference distribution of video descriptions. The target model on the matched pairs is learned on top the reference model, which improves the distinctiveness of generated captions. In addition, we further boost the distinctiveness of the captions by developing a hard mining technique to select the hardest mismatched pairs within the contrastive learning framework. Finally, the relationships among multiple relevant captions for each video is consider to encourage the diversity of generated captions. The proposed method generates high quality captions which effectively capture the specialties in individual videos. Extensive experiments on two benchmark datasets, i.e., MSVD and MSR-VTT, show that our approach outperforms state-of-the-art methods.
B-mode ultrasound tongue imaging is a non-invasive and real-time method for visualizing vocal tract deformation. However, accurately extracting the tongue’s surface contour remains a significant challenge due to the ...
详细信息
ISBN:
(数字)9798350368741
ISBN:
(纸本)9798350368758
B-mode ultrasound tongue imaging is a non-invasive and real-time method for visualizing vocal tract deformation. However, accurately extracting the tongue’s surface contour remains a significant challenge due to the low signal-to-noise ratio (SNR) and prevalent speckle noise in ultrasound images. Traditional supervised learning models often require large labeled datasets, which are labor-intensive to produce and susceptible to noise interference. To address these limitations, we present a novel Counterfactual Ultrasound Anti-Interference Self-Supervised Network (CUAI-SSN), which integrates self-supervised learning (SSL) with counterfactual data augmentation, progressively disentangles confounding factors, ensuring that the model generalizes well across varied ultrasound conditions. Our approach leverages causal reasoning to decouple noise from relevant features, enabling the model to learn robust representations that focus on essential tongue structures. By generating counterfactual image-label pairs, our method introduces alternative, noise-independent scenarios that enhance model training. Furthermore, we introduce attention mechanisms to enhance the network’s ability to capture fine-grained details even in noisy conditions. Extensive experiments on real ultrasound tongue images demonstrate that CUAI-SSN outperforms existing methods, setting a new benchmark for automated contour extraction in ultrasound tongue imaging. Our code is publicly available at https://***/inexhaustible419/CounterfactualultrasoundAI.
Basic recursive summation and common dot product algorithm have a backward error bound that grows linearly with the vector dimension. Blanchard [1] proposed a class of fast and accurate summation and dot product algor...
详细信息
The scale of model parameters and the amount of training data is exponentially increasing. It requires more GPU memory with the exponential increasement of model parameters. Recomputation and swapping are two main mem...
The scale of model parameters and the amount of training data is exponentially increasing. It requires more GPU memory with the exponential increasement of model parameters. Recomputation and swapping are two main memory optimization methods that have been extensively studied, and there are also optimization strategies that combine the two methods. However, most of them are based on heuristic search strategies, which do not explore the complete solution space and can’t guarantee the optimality of the solution results. An optimal search strategy with tensor-level recomputation and swapping is expected in large-scale model training. In this paper, we propose an optimal strategy searching algorithm combining tensor-based recomputation and swapping. Specifically, the memory swapping strategy is reformulated as an optimization problem, which converts the memory constraints into mixed integer programming, to find the optimal memory optimization strategy. By leveraging the advantages of both recomputation and swapping, this approach minimizes computation consumption without exceeding the available memory limitation. Experimental results show that our method exhibits about 60% reduction in memory requirements during the training process. Furthermore, our method can reduce the overall training time beyond the existing algorithms. Compared to Checkmate, our approach achieves about 0.3–0.9% reduction in computation cost per iteration.
Unlike Emotion Cause Extraction (ECE) task which consists of pre-annotate emotions and passage, emotion-cause pair extraction (ECPE) aims at extracting potential emotions and corresponding causes in the document witho...
详细信息
Accurate and efficient airway segmentation is essential for evaluating pulmonary diseases, aiding diagnosis, reducing the preoperative burden of airway identification, and minimizing patient discomfort during prolonge...
详细信息
ISBN:
(数字)9798350386226
ISBN:
(纸本)9798350386233
Accurate and efficient airway segmentation is essential for evaluating pulmonary diseases, aiding diagnosis, reducing the preoperative burden of airway identification, and minimizing patient discomfort during prolonged surgeries. However, current pulmonary airway reconstruction techniques are hindered by two major challenges: difficulty in accurately reconstructing fine airway branches due to the tendency to overlook small targets, and insufficient structural connectivity leading to frequent branch discontinuities within the airway tree. These limitations directly affect the clinical applicability of reconstructed airways. To overcome these challenges, a novel 3D pulmonary airway segmentation multi-task framework is proposed, designed to enhance the performance of existing backbone models. This approach integrates Anatomical Prior-Based Multi-Task Learning (AP-MTL) through the use of Gaussian-constructed connectivity-enhanced isosurfaces, significantly improving the network’s ability to maintain airway continuity. Additionally, a Class-Balanced CT Density Distribution Reconstruction mechanism (DDR-CB) is introduced, further refining the model’s capability to detect and segment fine airway branches. As a result of these enhancements, the model demonstrates a 11.5% average improvement in segmentation accuracy and connectivity compared to the baseline. The source code is publicly accessible at https://***/inexhaustible419/APMTLAirwaySegment.
Foundation models are in the process of becoming the dominant deep learning technology. Pretraining a foundation model is always time-consuming due to the large scale of both the model parameter and training dataset. ...
详细信息
Graph neural networks (GNNs) have been becoming important tools for processing structured graph data and successfully applied to multiple graph-based application scenarios. The existing GNN systems adopt sample-based ...
详细信息
Graph neural networks (GNNs) have been becoming important tools for processing structured graph data and successfully applied to multiple graph-based application scenarios. The existing GNN systems adopt sample-based training on large-scale graphs over multiple GPUs. Although they support large-scale graph training, large data loading overhead of transferring vertex features between CPUs and GPUs is still a bottleneck. In this work, we propose SCGraph, a method that supports GPU high-speed feature caching. SCGraph classifies the graph vertices sorted by out-degrees. For high out-degree vertices, SCGraph sets grading caches via different GPUs to increase the overall cache content through NVLink high-speed data transmission between them. For low out-degree vertices, SCGraph expands training vertices' neighborhood in advance to regenerate cache. We evaluate SCGraph against two state-of-the-art industrial GNN frameworks, i.e., DGL and PaGraph on various benchmarks. Experimental results show that SCGraph improves the cache hit rate over GPUs up to 23.6%, and achieves up to 1.71x performance speedup over the state-of-the-art baselines while the convergence almost constant.
暂无评论