This paper provides a comprehensive overview of artificial neuralnetworks (ANNs), exploring their theoretical foundations, practical applications, and recent advancements. I delve into the basic constructs of ANNs, d...
详细信息
Type 2 diabetes (T2D) is a prolonged disease caused by abnormal rise in glucose levels due to poor insulin production in the pancreas. However, the detection and classification of this type of disease is very challeng...
详细信息
Deep neuralnetwork (DNN) models are becoming ubiquitous in a variety of contemporary domains such as Autonomous Vehicles, Smart cities and Healthcare. They help drones to navigate, identify suspicious activities from...
详细信息
In this study, we propose BreathPass, a non-invasive authentication system that characterizes the chest/abdomen movement incurred by human breath to enable unlocking smart devices while wearing various types of face c...
详细信息
Deep neuralnetwork (DNN) applications are pervasive. However as demands for these applications continue to increase, so is the challenges for designing flexible and scalable architectures for multi-application implem...
详细信息
ISBN:
(纸本)9798350323481
Deep neuralnetwork (DNN) applications are pervasive. However as demands for these applications continue to increase, so is the challenges for designing flexible and scalable architectures for multi-application implementation. Such accelerators require innovative architecture with flexible network-on-Chips (NoCs), parallelism exploitation, and better on-chip memory organization to adequately support the diverse computation, memory, and communication needs. In this paper, we propose Venus, a versatile DNN accelerator design that can provide efficient communication and computation support for multi-applications. Venus is a tile-based architecture with a distributed buffer where each tile consists of an array of processing elements (PEs) and a portion of the distributed buffer. The other salient feature of Venus is a flexible network-on-Chip (NoC) that can dynamically adapt to the communication needs of various running applications thus maximizing data reuse, reducing DRAM accesses, and supporting multiple dataflows with an overall aim of better execution time and better energy efficiency. Simulation results show that our proposed Venus design outperforms state-of-art accelerators (NVDLA [1], ShiDianNao [2], Eyeriss [3], Planaria [4], Simba [5]).
Research in the autonomous underwater detection system has become rapidly increasing in Ocean Technology. In a recent object detection research study, there a need to enhance the quality, which needs to handle submerg...
详细信息
Research in the autonomous underwater detection system has become rapidly increasing in Ocean Technology. In a recent object detection research study, there a need to enhance the quality, which needs to handle submerged object image processing techniques and a lot of demand to develop an intelligent vision system to improve the Blurred Images and low-quality illumination. Manual research in undersea water leads to more significant pressures and complex environments in cost and workforce. It is necessary to develop a high acceptable autonomous image quality system to upgrade image quality. This paper proposed two approaches: (i) Gray shade and Max-RGB filter techniques to improve image quality. (ii) For optimization and low illumination problem modified Convolution neural Technique (CNN) incorporated for classification and detection. Moreover, our proposed model has compared with Single-shot Detector (SDD), You Only Look Once (Yolo), Fast RCNN, Faster RCNN to uphold the quality detection found objects. This research article aids to found real-time underwater objects classification and detection. It helps to incorporate an Autonomous operation Vehicle (AOV) underwater research. Our experiment results show detection runs speed as 30 FPs (Frame per second).
Nowadays, data analysis is widely used in numerous areas to identify new trends, opportunities, or risks and to improve decision-making. In many cases, however, data analysis is only possible by incorporating specific...
详细信息
ISBN:
(纸本)9783031647475;9783031647482
Nowadays, data analysis is widely used in numerous areas to identify new trends, opportunities, or risks and to improve decision-making. In many cases, however, data analysis is only possible by incorporating specific domain knowledge, which is why domain experts need to be involved. To this end, data mashups are a popular tool for modeling tailored analyses. Yet, with today's data volumes from heterogeneous source systems, it is very difficult to identify beneficial data sources, in particular for explorative data analysis. In this paper, we first define requirements aiming for user-centric analytics, followed by the introduction of SDRank, a deep-learning-based approach to identify beneficial data sources. In an extensive evaluation with three scenarios, we show that this approach offers high robustness concerning the training data used and can reliably identify beneficial data sources, even for previously unknown domains, i.e., transfer learning.
Graph neuralnetworks (GNNs) have emerged as a popular choice for analyzing structured data organized as graphs. Nevertheless, GNN models tend to be shallow, failing to fully exploit the capabilities of modern GPUs. O...
详细信息
ISBN:
(纸本)9783031697654;9783031697661
Graph neuralnetworks (GNNs) have emerged as a popular choice for analyzing structured data organized as graphs. Nevertheless, GNN models tend to be shallow, failing to fully exploit the capabilities of modern GPUs. Our motivational tests reveal that GPU dataloading for GNN inference yields remarkable performance enhancements when both the graph topology and features reside in GPU memory. Unfortunately, the use of this approach is hindered by the large size of real-world graph datasets. To address this limitation, we introduce GDL-GNN, a partition-based method that incorporates all essential information for inference within each subgraph. It thus combines the efficiency of GPU dataloading with layerwise inference, while maintaining the accuracy of full-neighbor inference. Additional optimization enables GDL-GNN to avoid unnecessary representation computation on halo nodes and to conceal file loading time. Evaluation shows the effectiveness of GDL-GNN in both single- and multi-GPU scenarios, revealing a reduction in inference time of up to 59.9% without compromising accuracy.
The rapid evolution of artificial intelligence (AI), especially in deep learning, is significantly driven by big data. However, the intensive resources required for training deep neuralnetworks (DNN) highlight the ur...
详细信息
ISBN:
(纸本)9783031697654;9783031697661
The rapid evolution of artificial intelligence (AI), especially in deep learning, is significantly driven by big data. However, the intensive resources required for training deep neuralnetworks (DNN) highlight the urgent need for effective model protection and ownership verification. Current neuralnetwork watermarking methods fall short in federated learning contexts. This paper introduces VeriChroma, an innovative framework crafted to secure DNN models and affirm ownership within such environments. VeriChroma enables clients to embed and verify private ID-based watermarks independently, ensuring straightforward ownership claims. Through strategic image blocking and positional mapping, it overcomes conflicts between client constraints, ensuring tailored watermark integration. Furthermore, VeriChroma utilizes RGB filters for watermark triggers, bolstering both the robustness and stealth of the watermarking process. Our findings underscore VeriChroma's effectiveness and practicality, showcasing its potential to enhance DNN model security, resolve federated learning disputes, and provide secure, unobtrusive watermarking, marking a significant advancement in federated learning security and intellectual property rights protection.
COVID-19 pandemic has significantly changed learning processes. Learning, which had generally been carried out face-to-face, has now turned online. This learning strategy has both advantages and challenges. On the bri...
详细信息
COVID-19 pandemic has significantly changed learning processes. Learning, which had generally been carried out face-to-face, has now turned online. This learning strategy has both advantages and challenges. On the bright side, online learning is unbound by space and time, allowing it to take place anywhere and anytime. On the other side, it faces a common challenge in the lack of direct interaction between educators and students, making it difficult to assess students' engagement during an online learning process. Therefore, it is necessary to conduct research with the aim of automatically detecting students' engagement during online learning. The data used in this research were derived from the DAiSEE dataset (Dataset for Affective States in E-Environments), which comprises ten-second video recordings of students. This dataset classifies engagement levels into four categories: low, very low, high, and very high. However, the issue of imbalanced data found in the DAiSEE dataset has yet to be addressed in previous research. This data imbalance can cause errors in the classification model, resulting in overfitting and underfitting of the model. In this study, Convolutional neuralnetwork, a deep learning model, was utilized for feature extraction on the DAiSEE dataset. The OpenFace library was used to perform facial landmark detection, head pose estimation, facial expression unit recognition, and eye gaze estimation. The pre-processing stages included data selection, dimensional reduction, and normalization. The PCA and SVD techniques were used for dimensional reduction. The data were later oversampled using the SMOTE algorithm. The training and testing data were distributed at an 80:20 ratio. The results obtained from this experiment exceeded the benchmark evaluation values on the DAiSEE dataset, achieving the best accuracy of 77.97% using the SVD dimensional reduction technique.
暂无评论