Blockchain-Based Distributed Deep Learning Task Assignment Scheme

In industry deep learning application, our manually labeled data has a certain number of noisy data. To solve this problem and achieve more than 90 score in dev dataset, we present a simple method to find the noisy data and re-label the noisy data by human, given the model predictions as references in human labeling. In this paper, we illustrate our idea for a broad set of deep learning tasks, includes classification, sequence tagging, object detection, sequence generation, click-through rate prediction. The experimental results and human evaluation results verify our idea.

Download Full-text

Modelling a Deep Learning-Based Wireless Sensor Network Task Assignment Algorithm

Deep Learning Strategies for Security Enhancement in Wireless Sensor Networks - Advances in Information Security, Privacy, and Ethics ◽

10.4018/978-1-7998-5068-7.ch005 ◽

2020 ◽

pp. 84-109

Author(s):

Titus Issac ◽

Salaja Silas ◽

Elijah Blessing Rajsingh

Keyword(s):

Deep Learning ◽

Wireless Sensor Network ◽

Sensor Network ◽

Task Assignment ◽

Human Interaction ◽

Sensor Nodes ◽

Wireless Sensor ◽

Feed Forward Neural Network ◽

Functional Capabilities ◽

Assignment Algorithm

The 21st century is witnessing the emergence of a wide variety of wireless sensor network (WSN) applications ranging from simple environmental monitoring to complex satellite monitoring applications. The advent of complex WSN applications has led to a massive transition in the development, functioning, and capabilities of wireless sensor nodes. The contemporary nodes have multi-functional capabilities enabling the heterogeneous WSN applications. The future of WSN task assignment envisions WSN to be heterogeneous network with minimal human interaction. This led to the investigative model of a deep learning-based task assignment algorithm. The algorithm employs a multilayer feed forward neural network (MLFFNN) trained by particle swarm optimization (PSO) for solving task assignment problem in a dynamic centralized heterogeneous WSN. The analyses include the study of hidden layers and effectiveness of the task assignment algorithms. The chapter would be highly beneficial to a wide range of audiences employing the machine and deep learning in WSN.

Download Full-text

Task Assignment Scheme of Multi-UAV Based on Single-Unit Combinatorial Auction

2010 International Conference on Electrical and Control Engineering ◽

10.1109/icece.2010.550 ◽

2010 ◽

Author(s):

Zhu Mingjian ◽

Qin Zheng ◽

Xing Jiankuan

Keyword(s):

Single Unit ◽

Task Assignment ◽

Combinatorial Auction ◽

Assignment Scheme ◽

Multi Uav

Download Full-text

Research on Multi-objective Task Assignment Scheme Based on Group Isomorphism UUV

10.1109/iccsse52761.2021.9545203 ◽

2021 ◽

Author(s):

Ke YongSheng ◽

Wu Rui ◽

Guo Xuan ◽

Luo GuangYu

Keyword(s):

Task Assignment ◽

Group Isomorphism ◽

Multi Objective ◽

Assignment Scheme

Download Full-text

Deep Learning Accelerators: A Case Study

10.21203/rs.3.rs-24147/v1 ◽

2020 ◽

Author(s):

Hamidreza Bolhasani ◽

Somayyeh Jafarali Jassbi

Keyword(s):

Deep Learning ◽

Language Processing ◽

Performance Indicators ◽

Learning Task ◽

High Accuracy ◽

The Other ◽

Almost All ◽

L1 And L2 ◽

Accuracy Speed

Abstract In the recent years, deep learning has become one of the most important topics in computer science. Deep learning is a growing trend in the edge of technology and its applications are now seen in many aspects of our life such as object detection, speech recognition, natural language processing, etc. Currently, almost all major sciences and technologies are benefiting from the advantages of deep learning such as high accuracy, speed and flexibility. Therefore, any efforts for improving performance of related techniques is valuable. Deep learning accelerators are considered as hardware architecture, which are designed and optimized for increasing the speed, efficiency and accuracy of computers that are running deep learning algorithms. In this paper, after reviewing some backgrounds about deep learning, a well-known accelerator architecture named MAERI is investigated. By using an open source tool called MAESTRO, the performance of a deep learning task is measured and compared on two different data flow strategies: NLR and NVDLA. Measured performance indicators of novel optimized architecture, NVDLA shows higher L1 and L2 computation reuse and lower total runtime (cycles) in comparison to the other one.

Download Full-text

Stock Price Prediction using Technical Indicators: A Predictive Model using Optimal Deep Learning

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.b3048.078219 ◽

2019 ◽

Vol 8 (2) ◽

pp. 2297-2305

Keyword(s):

Deep Learning ◽

Stock Market ◽

Stock Price ◽

Short Term Memory ◽

Technical Analysis ◽

Learning Task ◽

Correlation Tensor ◽

Technical Indicators ◽

Stock Price Prediction ◽

Proposed Model

The stock market is highly volatile and complex in nature. Technical analysts often apply Technical Analysis (TA) on historical price data, which is an exhaustive task and might produce incorrect predictions. The machine learning coupled with fundamental and / or Technical Analysis also yields satisfactory results for stock market prediction. In this work an effort is made to predict the price and price trend of stocks by applying optimal Long Short Term Memory (O-LSTM) deep learning and adaptive Stock Technical Indicators (STIs). We also evaluated the model for taking buy-sell decision at the end of day. To optimize the deep learning task we utilized the concept of Correlation-Tensor built with appropriate STIs. The tensor with adaptive indicators is passed to the model for better and accurate prediction. The results are analyzed using popular metrics and compared with two benchmark ML classifiers and a recent classifier based on deep learning. The mean prediction accuracy achieved using proposed model is 59.25%, over number of stocks, which is much higher than benchmark approaches.

Download Full-text

MFNet algorithm based on indoor scene segmentation

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-212275 ◽

2021 ◽

pp. 1-10

Author(s):

Rui Cao ◽

Feng Jiang ◽

Zhao Wu ◽

Jia Ren

Keyword(s):

Deep Learning ◽

Learning Task ◽

Vital Role ◽

Scene Segmentation ◽

Feature Maps ◽

Real Time Processing ◽

Multi Scale ◽

Indoor Scene ◽

Convolution Kernels ◽

Hardware Platforms

With the advancement of computer performance, deep learning is playing a vital role on hardware platforms. Indoor scene segmentation is a challenging deep learning task because indoor objects tend to obscure each other, and the dense layout increases the difficulty of segmentation. Still, current networks pursue accuracy improvement, sacrifice speed, and augment memory resource usage. To solve this problem, achieve a compromise between accuracy, speed, and model size. This paper proposes Multichannel Fusion Network (MFNet) for indoor scene segmentation, which mainly consists of Dense Residual Module(DRM) and Multi-scale Feature Extraction Module(MFEM). MFEM uses depthwise separable convolution to cut the number of parameters, matches different sizes of convolution kernels and dilation rates to achieve optimal receptive field; DRM fuses feature maps at several levels of resolution to optimize segmentation details. Experimental results on the NYU V2 dataset show that the proposed method achieves very competitive results compared with other advanced algorithms, with a segmentation speed of 38.47 fps, nearly twice that of Deeplab v3+, but only 1/5 of the number of parameters of Deeplab v3 + . Its segmentation results were close to those of advanced segmentation networks, making it beneficial for the real-time processing of images.

Download Full-text