Operators(such as Conv and ReLU) play an important role in deep neural networks. Every neural network is composed of a series of differentiable operators. However, existing AI benchmarks mainly focus on accessing mode...
详细信息
Operators(such as Conv and ReLU) play an important role in deep neural networks. Every neural network is composed of a series of differentiable operators. However, existing AI benchmarks mainly focus on accessing model training and inference performance of deep learning systems on specific models. To help GPU hardware find computing bottlenecks and intuitively evaluate GPU performance on specific deep learning tasks, this paper focuses on evaluating GPU performance at the operator level. We statistically analyze the information of operators on 12 representative deep learning models from six prominent AI tasks and provide an operator dataset to show the different importance of various types of operators in different networks. An operator-level benchmark, OpBench, is proposed on the basis of this dataset, allowing users to choose from a given range of models and set the input sizes according to their demands. This benchmark offers a detailed operator-level performance report for AI and hardware developers. We also evaluate four GPU models on OpBench and find that their performances differ on various types of operators and are not fully consistent with the performance metric FLOPS(floating point operations per second).
Internet of Things (IoT) devices are often directly authenticated by the gateways within the network. In complex and large systems, IoT devices may be connected to the gateway through another device in the network. In...
详细信息
Federated Adversarial Learning (FAL) maintains the decentralization of adversarial training for data-driven innovations while allowing the collaborative training of a common model to protect privacy facilities. Before...
详细信息
In the times of advanced generative artificial intelligence, distinguishing truth from fallacy and deception has become a critical societal challenge. This research attempts to analyze the capabilities of large langua...
详细信息
The proliferation of Internet of Things(IoT)technology has exponentially increased the number of devices interconnected over networks,thereby escalating the potential vectors for cybersecurity *** response,this study ...
详细信息
The proliferation of Internet of Things(IoT)technology has exponentially increased the number of devices interconnected over networks,thereby escalating the potential vectors for cybersecurity *** response,this study rigorously applies and evaluates deep learning models—namely Convolutional Neural Networks(CNN),Autoencoders,and Long Short-Term Memory(LSTM)networks—to engineer an advanced Intrusion Detection System(IDS)specifically designed for IoT *** the comprehensive UNSW-NB15 dataset,which encompasses 49 distinct features representing varied network traffic characteristics,our methodology focused on meticulous data preprocessing including cleaning,normalization,and strategic feature selection to enhance model performance.A robust comparative analysis highlights the CNN model’s outstanding performance,achieving an accuracy of 99.89%,precision of 99.90%,recall of 99.88%,and an F1 score of 99.89%in binary classification tasks,outperforming other evaluated models *** results not only confirm the superior detection capabilities of CNNs in distinguishing between benign and malicious network activities but also illustrate the model’s effectiveness in multiclass classification tasks,addressing various attack vectors prevalent in IoT *** empirical findings from this research demonstrate deep learning’s transformative potential in fortifying network security infrastructures against sophisticated cyber threats,providing a scalable,high-performance solution that enhances security measures across increasingly complex IoT *** study’s outcomes are critical for security practitioners and researchers focusing on the next generation of cyber defense mechanisms,offering a data-driven foundation for future advancements in IoT security strategies.
Low-light image enhancement is highly desirable for outdoor image processing and computer vision applications. Research conducted in recent years has shown that images taken in low-light conditions often pose two main...
详细信息
In this study, tests were done to see what would happen if hydrogen (H2) and lemon grass oil (LO) were used for a lone-cylinder compression ignition engine as a partial diesel replacement. After starting the trial wit...
详细信息
Machine learning models have been prevalently deployed for malware detection. When properly trained under the training environment, they can deliver highly accurate detection results in the deployment environment prov...
详细信息
Machine learning models have been prevalently deployed for malware detection. When properly trained under the training environment, they can deliver highly accurate detection results in the deployment environment provided that the two environments have no distribution shift from each other. Yet, in practice, various factors can cause environment shifts, which lead to degraded malware detection accuracy. In this work, we propose SCRR, a unified training framework to enhance the stability of malware detection models under unknown distribution shifts of the deployment environment. Our method can enhance the stability of the model by filtering out correlations between malicious behaviours and irrelevant features, known as the SC (spurious correlation), which can change significantly across different environments. What is more, SCRR proposes a fine-grained SC filtering strategy to achieve better accuracy performance. We evaluate SCRR in terms of in-distribution accuracy, degradation under environment shifts, and comprehensive detection ability with two real-world Android malware datasets, considering three types of causal factors and four environment shifts. SCRR outperforms the state-of-the-art malware detection training methods by improving the detection accuracy by up to 13.4% under the considered environment shifts. Moreover, it consistently showcases in-distribution accuracy comparable to the best outcomes achieved by baseline methods. IEEE
Deformable image registration is a fundamental technique in medical image analysis and provide physicians with a more complete understanding of patient anatomy and function. Deformable image registration has potential...
详细信息
Gait recognition has a wide range of application scenarios in the fields of intelligent security and *** recognition currently faces challenges:inadequate feature methods for environmental interferences and insufficie...
详细信息
Gait recognition has a wide range of application scenarios in the fields of intelligent security and *** recognition currently faces challenges:inadequate feature methods for environmental interferences and insufficient local-global information *** address these issues,we propose a gait recognition model based on feature fusion and dual *** model utilizes the ResNet architecture as the backbone network for fundamental gait features ***,the features from different network layers are passed through the feature pyramid for feature fusion,so that multi-scale local information can be fused into global information,providing a more complete feature *** dual attention module enhances the fused features in multiple dimensions,enabling the model to capture information from different semantics and scale *** model proves effective and competitive results on CASIA-B(NM:95.6%,BG:90.9%,CL:73.7%)and OU-MVLP(88.1%).The results of related ablation experiments show that the model design is effective and has strong competitiveness.
暂无评论