With advancements in AI infrastructure and Trusted Execution Environment (TEE) technology, Federated Learning as a Service (FLaaS) through JointCloud Computing (JCC) is promising to break through the resource constrai...
详细信息
Multi-view learning has been explored for audio classification tasks, exploiting different representations of audio signals, ranging from MFCC, CQT, to raw signals. The quality of each view may vary for different audi...
Multi-view learning has been explored for audio classification tasks, exploiting different representations of audio signals, ranging from MFCC, CQT, to raw signals. The quality of each view may vary for different audio signals, and the appropriate uncertainty quantification for each view has not been fully explored. In this work, we explore a trusted multi-view learning framework for classification tasks in order to fully incorporate different views. Our framework consists of three parallel branches of Transformer architectures (Gammatone spectrogram, log-Mel and CQT) and they are combined using the uncertainty estimation of different branch. In addition to computing the classification probabilities, the uncertainty of each representation can also be obtained using the framework. We firstly calculate the evidence based on feature vectors to obtain the probabilities and the uncertainty of classification problems for Gammatone, log-Mel and CQT branch. By integrating the confidence from each of the different representations using the Dempster–Shafer theory, the classification framework can provide higher accuracy and confidence. To demonstrate the effectiveness of the proposed framework, we conduct the experiments on the GTZAN dataset. The obtained results show that our method can reach the accuracy of 83.0%, which significantly outperforms single representation-based methods while providing uncertainty estimation for different views.
Underwater acoustic classification is a challenging task due to complex background noise and complicated sound propagation patterns. How to represent the signals is important for the classification task. In this paper...
Underwater acoustic classification is a challenging task due to complex background noise and complicated sound propagation patterns. How to represent the signals is important for the classification task. In this paper, we propose a novel representation learning method for the underwater acoustic signals, leveraging the mask modeling-based self-supervised learning paradigm. Specifically, we first explore modifying the Swin Transformer architecture to learn general representation for the audio signals, accompanied with random masking on the log-mel spectrogram. The main goal of the pretext task is to predict the masked parts of Log-mel spectrogram and the gamma-stone spectrogram, so that the model can not only learn the local and global features but also learn complementary information. For downstream task, we utilize the lab.lled datasets to fine-tune the pre-trained model. On DeepShip datasets which consist of 47 hand 4 minof ship sounds in four categories, our model achieves state-of-the-art performance compared with competitive approaches. Our method obtains a classification accuracy of 78.03%, which is better than the separable convolution autoencoder (SCAE) and using the constant-Q transform spectrogram. This work demonstrates the potential of the masked modeling based self-supervised learning for understanding and interpretation of underwater acoustic signals.
This paper presents a load balancing method for a multi-block grids-based CFD (Computational Fluid Dynamics) application on heterogeneous platform. This method includes an asymmetric task scheduling scheme and a load ...
详细信息
ISBN:
(数字)9781665403986
ISBN:
(纸本)9781665403993
This paper presents a load balancing method for a multi-block grids-based CFD (Computational Fluid Dynamics) application on heterogeneous platform. This method includes an asymmetric task scheduling scheme and a load balancing model. The idea is to balance the computing speed between the CPU and the coprocessor by adjusting the workload and the numbers of threads on both sides. Optimal load balance parameters are empirically selected, guided by a performance model. Performance evaluation is conducted on a computer server consists of two Intel Xeon E5-2670 v3 CPUs and two MIC coprocessors (Xeon Phi 5110P and Xeon Phi 7120P) for the simulation of turbulent combustion in a supersonic combustor. The results show that the performance is highly sensitive to the load balance parameters. With the optimal parameters, the heterogeneous computing achieves a maximum speedup of 2.30 × for a 6-block mesh, and a maximum speedup of 2.66 × for a 8-block mesh, over the CPU-only computing.
The feasibility of age estimation is explored using the ultrasound tongue image of the speakers. Motivated by the success of deep learning, a deep convolutional neural network model is trained on the UltraSuite datase...
The feasibility of age estimation is explored using the ultrasound tongue image of the speakers. Motivated by the success of deep learning, a deep convolutional neural network model is trained on the UltraSuite dataset. The deep model achieves mean absolute error (MAE) of 2.03 years for the data from typically developing children, while MAE is 4.87 for the data from the children with speech sound disorders, which suggest that age estimation using ultrasound is more challenging for the children with speech sound disorder. Also, we explore to visualize what does the deep model learn for the age estimation task. We firstly visualize the convolutional layers in the learned convolutional neural networks. We observe that the deep model not only focuses on the contour in the ultrasound tongue image, but also pays more attention to the regions corresponding to the tendon and tongue root regions, which may provide guidance for future ultrasound tongue imaging interpretation tasks. The developed method can be used a tool to evaluate the performance of speech therapy sessions.
Last-Level Cache (LLC) plays an important role in Chip Multi-Processor (CMP). The objective of this work is to optimize the structure and management strategy of LLC. Based on 8-core CMP, a LLC structure based on group...
详细信息
ISBN:
(纸本)9781479975761
Last-Level Cache (LLC) plays an important role in Chip Multi-Processor (CMP). The objective of this work is to optimize the structure and management strategy of LLC. Based on 8-core CMP, a LLC structure based on grouped cores is proposed, where 8 cores are divided into 4 groups. All LLC resources are classified into three types, which are fixed private cache, dynamic private cache and dynamic shared cache. The layout of the LLC structure and the corresponding dynamic partitioning strategy are designed to achieve low access latency and high efficiency. Experimental results on full-system simulator suggest that the proposed structure and method are able to reduce the access latency by 2% to 12% compared with previous works, such as tiled structure, cache-centered structure and core-centered structure. Consequently, performance measured by IPC is improved up to 7%. The contribution of this paper is useful for CMP performance, and applies to not only 8-core CMP but also all small-scale CMPs.
Many recent applications involve processing and analyzing uncertain data. Recently, several research efforts have addressed answering skyline queries efficiently on massive uncertain datasets. However, the research la...
详细信息
By recognizing the necessity for preventative and proactive management for today's large scale and fault prone distributed systems, a tendency for these mechanisms has been appeared in recent researchers' effo...
详细信息
Nowadays by improving the richness of prediction methods and accessing to the more information about systems behavior, the role of proactive strategies in developing more reliable and efficient systems becomes more cr...
详细信息
The Sparse Matrix-Vector product (SpMV) is a key operation in engineering and scientific computing. Methods for efficiently implementing it in parallel are critical to the performance of many applications. Modern Grap...
详细信息
暂无评论