BAIPAS: Distributed Deep Learning Platform with Data Locality and Shuffling

Distributed deep learning is a promising approach for training and deploying large and complex deep learning models. This paper presents a comprehensive workflow for deploying and optimizing the YOLACT instance segmentation model as on big data clusters. OpenVINO, a toolkit known for its high-speed data processing and ability to optimize deep learning models for deployment on a variety of devices, was used to optimize the YOLACT model. The model is then run on a big data cluster using BigDL, a distributed deep learning library for Apache Spark. BigDL provides a high-level programming interface for defining and training deep neural networks, making it suitable for large-scale deep learning applications. In distributed deep learning, input data is divided and distributed across multiple machines for parallel processing. This approach offers several advantages, including the ability to handle very large data that can be stored in a distributed manner, scalability to decrease processing time by increasing the number of workers, and fault tolerance. The proposed workflow was evaluated on virtual machines and Azure Databricks, a cloud-based platform for big data analytics. The results indicated that the workflow can scale to large datasets and deliver high performance on Azure Databricks. This study explores the benefits and challenges of using distributed deep learning on big data clusters for instance segmentation. Popular distributed deep learning frameworks are discussed, and BigDL is chosen. Overall, this study highlights the practicality of distributed deep learning for deploying and scaling sophisticated deep learning models on big data clusters.

Using big marine data to train deep learning models is not efficient, or sometimes even possible, on local computers. In this paper, we show how distributed learning in the cloud can help more efficiently process big data and train more accurate deep learning models. In addition, marine big data is usually communicated over wired networks, which if possible to deploy in the first place, are costly to maintain. Therefore, wireless communications dominantly conducted by acoustic waves in underwater sensor networks, may be considered. However, wireless communication is not feasible for big marine data due to the narrow frequency bandwidth of acoustic waves and the ambient noise. To address this problem, we propose an optimized deep learning design for low-energy and real-time image processing at the underwater edge. This leads to trading the need to transmit the large image data, for transmitting only the low-volume results that can be sent over wireless sensor networks. To demonstrate the benefits of our approaches in a real-world application, we perform fish segmentation in underwater videos and draw comparisons against conventional techniques. We show that, when underwater captured images are processed at the collection edge, 4 times speedup can be achieved compared to using a landside server. Furthermore, we demonstrate that deploying a compressed DNN at the edge can save 60% of power compared to a full DNN model. These results promise improved applications of affordable deep learning in underwater exploration, monitoring, navigation, tracking, disaster prevention, and scientific data collection projects.

Stochastic gradient boosting

Nowadays, big data is directing the entire advanced world with its function and applications. Moreover, to make better decisions from the ever emerging big data belonging to the respective organizations, deep learning (DL) models are required. DL is also widely used in the sentiment classification tasks considering data from social networks. Furthermore, sentiment classification signifies the best way to analyze the big data and make decisions accordingly. Analyzing the sentiments from big data applications is quite challenging task and also requires more time for the execution process. Therefore, to analyze and classify big data emerging from social networks in a better way, DL models are utilized. DL techniques are being used among the researchers to get high end results. A novel Ant Colony-based Deep Belief Neural Network (AC-DBN) framework is proposed in this research. Drug review tweets are opted to perform sentiment classification by using the proposed framework in python environment. A model fitness function is initiated in the DL framework and is observed that it is attaining high accuracy with low computation time. Additionally, the obtained results attained from the proposed framework are validated with existing methods for evaluating the efficiency of the proposed AC-DBN approach.

The large-scale model training is typically slower and necessitates high-performance workstations and Distributed Deep Learning (DDL). The DDL models trained on a massive volume of data can outperform single accelerators being used, i.e., the performance of deep learning (DL) models can be enhanced by using distributed and parallel deep learning methods. But the DDL models need to be redesigned and evaluated for particular hardware resources and applications, requiring the literature on various DDL model evaluations for assistance. Authors in the proposed work design and evaluate DDL with data parallelism for classifying Diabetic retinopathy (DR) images. A performance comparative study on the DDL strategy implemented on DenseNet is presented where the model is trained on multi-GPUs with counts from 1 to 8. The results show that the model takes less time when trained on multi-GPUs compared to single GPU. The detailed evaluation analysis of DDL with data parallelism is presented in this paper.

Over the last decade, tourism industry is exploring the latest technologies to improve the customer experience and enhance customer satisfaction. In order to address the customer demands and business, tourism industry embraces to redefine their products and services using deep learning models and big data analytics. Deep learning-based model can be used to measure customer satisfaction, perception and behavior utilizing sentiment analysis, emotion analysis, and data analysis. Big data analytics in tourism can improve the overall tourism business operations and services. Deep learning techniques with big data analytics not only helps in developing the economical tourism models but also analyze the impact of various environmental factors on tourism planning and travel demand. In addition, these techniques are helpful in estimation of tourist seasonal demands, market price strategies and data analysis automation. Deep learning using data analytics can bring a paradigm shift in the tourism industry by offering personalized recommendations with budget specific packages based on customer past travel, reviews and experience. In this chapter, we will briefly analyze the impact of deep learning in tourism sector in terms of innovation, commercialization and profitability of the business. In addition, we will also review the advancement and potential of the deep learning-based methods with big data analytics in tourism industry in terms of customer overall experience.

8 - Deep learning techniques for optimizing medical big data

Artificial intelligence (AI)-based approaches are crucial in computer-aided diagnosis (CAD) for various medical applications. Their ability to quickly and accurately learn from complex data is remarkable. Deep learning (DL) models have shown promising results in accurately classifying Alzheimer's disease (AD) and its related cognitive states, Early Mild Cognitive Impairment (EMCI) and Late Mild Cognitive Impairment (LMCI), along with the healthy conditions known as Cognitively Normal (CN). This offers valuable insights into disease progression and diagnosis. However, certain traditional machine learning (ML) classifiers perform equally well or even better than DL models, requiring less training data. This is particularly valuable in CAD in situations with limited labeled datasets. In this paper, we propose an ensemble classifier based on ML models for magnetic resonance imaging (MRI) data, which achieved an impressive accuracy of 96.52%. This represents a 3-5% improvement over the best individual classifier. We evaluated popular ML classifiers for AD classification under both data-scarce and data-rich conditions using the Alzheimer's Disease Neuroimaging Initiative and Open Access Series of Imaging Studies datasets. By comparing the results to state-of-the-art CNN-centric DL algorithms, we gain insights into the strengths and weaknesses of each approach. This work will help users to select the most suitable algorithm for AD classification based on data availability.

Inspired by the successful applications in image recognition, the very deep convolutional residual network (ResNet) based model has been applied in automatic speech recognition (ASR). However, the computational load is heavy for training the ResNet with a large quantity of data. In this paper, we propose an incremental model training framework to accelerate the training process of the ResNet. The incremental model training framework is based on the unequal importance of each layer and connection in the ResNet. The modules with important layers and connections are regarded as a skeleton model, while those left are regarded as an auxiliary model. The total depth of the skeleton model is quite shallow compared to the very deep full network. In our incremental training, the skeleton model is first trained with the full training data set. Other layers and connections belonging to the auxiliary model are gradually attached to the skeleton model and tuned. Our experiments showed that the proposed incremental training obtained comparable performances and faster training speed compared with the model training as a whole without consideration of the different importance of each layer.

Research in many fields has shown that transfer learning (TL) is well-suited to improve the performance of deep learning (DL) models in datasets with small numbers of samples. This empirical success has triggered interest in the application of TL to cognitive decoding analyses with functional neuroimaging data. Here, we systematically evaluate TL for the application of DL models to the decoding of cognitive states (e.g., viewing images of faces or houses) from whole-brain functional Magnetic Resonance Imaging (fMRI) data. We first pre-train two DL architectures on a large, public fMRI dataset and subsequently evaluate their performance in an independent experimental task and a fully independent dataset. The pre-trained DL models consistently achieve higher decoding accuracies and generally require less training time and data than model variants that were not pre-trained, while also outperforming linear baseline models trained from scratch, clearly underlining the benefits of pre-training. We demonstrate that these benefits arise from the ability of the pre-trained models to reuse many of their learned features when training with new data, providing deeper insights into the mechanisms giving rise to the benefits of pre-training. Yet, we also surface nuanced challenges for whole-brain cognitive decoding with DL models when interpreting the decoding decisions of the pre-trained models, as these have learned to utilize the fMRI data in unforeseen and counterintuitive ways to identify individual cognitive states.

DPro-SM – A distributed framework for proactive straggler mitigation using LSTM

Nowadays, large and complex deep learning (DL) models are increasingly trained in a distributed manner across multiple worker machines, in which extensive communications between workers pose serious scaling problems. In this article, we present a quantitative survey of communication optimization techniques for data parallel distributed DL. We first identify the major communication challenges and classify the existing solutions into three levels, namely the learning algorithm, the system architecture, and the network infrastructure. We present the state-of-the-art communication optimization techniques and conduct a comparative study of seven common lossless distributed DL methods on a 32-GPU cluster with 100Gb/s InfiniBand (IB). We show that the DL models with low model intensity (such as BERT and BERT-Large) are difficult to scale out even with the best available lossless algorithm over 100Gb/s IB; and the system architecture and scheduling algorithms have a critical impact on the scaling property. We conclude the article with discussions of open issues for further investigation.

COVID-19 is highly infectious and causes acute respiratory disease. Machine learning (ML) and deep learning (DL) models are vital in detecting disease from computerized chest tomography (CT) scans. The DL models outperformed the ML models. For COVID-19 detection from CT scan images, DL models are used as end-to-end models. Thus, the performance of the model is evaluated for the quality of the extracted feature and classification accuracy. There are four contributions included in this work. First, this research is motivated by studying the quality of the extracted feature from the DL by feeding these extracted to an ML model. In other words, we proposed comparing the end-to-end DL model performance against the approach of using DL for feature extraction and ML for the classification of COVID-19 CT scan images. Second, we proposed studying the effect of fusing extracted features from image descriptors, e.g., Scale-Invariant Feature Transform (SIFT), with extracted features from DL models. Third, we proposed a new Convolutional Neural Network (CNN) to be trained from scratch and then compared to the deep transfer learning on the same classification problem. Finally, we studied the performance gap between classic ML models against ensemble learning models. The proposed framework is evaluated using a CT dataset, where the obtained results are evaluated using five different metrics The obtained results revealed that using the proposed CNN model is better than using the well-known DL model for the purpose of feature extraction. Moreover, using a DL model for feature extraction and an ML model for the classification task achieved better results in comparison to using an end-to-end DL model for detecting COVID-19 CT scan images. Of note, the accuracy rate of the former method improved by using ensemble learning models instead of the classic ML models. The proposed method achieved the best accuracy rate of 99.39%.

PurposeThe purpose of this paper is to propose a distributed deep learning architecture for smart cities in big data systems.Design/methodology/approachWe have proposed an architectural multilayer to describe the distributed deep learning for smart cities in big data systems. The components of our system are Smart city layer, big data layer, and deep learning layer. The Smart city layer responsible for the question of Smart city components, its Internet of things, sensors and effectors, and its integration in the system, big data layer concerns data characteristics 10, and its distribution over the system. The deep learning layer is the model of our system. It is responsible for data analysis.FindingsWe apply our proposed architecture in a Smart environment and Smart energy. 10; In a Smart environment, we study the Toluene forecasting in Madrid Smart city. For Smart energy, we study wind energy foresting in Australia. Our proposed architecture can reduce the time of execution and improve the deep learning model, such as Long Term Short Memory10;.Research limitations/implicationsThis research needs the application of other deep learning models, such as convolution neuronal network and autoencoder.Practical implicationsFindings of the research will be helpful in Smart city architecture. It can provide a clear view into a Smart city, data storage, and data analysis. The 10; Toluene forecasting in a Smart environment can help the decision-maker to ensure environmental safety. The Smart energy of our proposed model can give a clear prediction of power generation.Originality/valueThe findings of this study are expected to contribute valuable information to decision-makers for a better understanding of the key to Smart city architecture. Its relation with data storage, processing, and data analysis.

Electricity load forecasting is a crucial undertaking within all the deregulated markets globally. Among the research challenges on a global scale, the investigation of deep transfer learning (DTL) in the field of electricity load forecasting represents a fundamental effort that can inform artificial intelligence applications in general. In this paper, a comprehensive study is reported regarding day-ahead electricity load forecasting. For this purpose, three sequence-to-sequence (Seq2seq) deep learning (DL) models are used, namely the multilayer perceptron (MLP), the convolutional neural network (CNN) and the ensemble learning model (ELM), which consists of the weighted combination of the outputs of MLP and CNN models. Also, the study focuses on the development of different forecasting strategies based on DTL, emphasizing the way the datasets are trained and fine-tuned for higher forecasting accuracy. In order to implement the forecasting strategies using deep learning models, load datasets from three Greek islands, Rhodes, Lesvos, and Chios, are used. The main purpose is to apply DTL for day-ahead predictions (1–24 h) for each month of the year for the Chios dataset after training and fine-tuning the models using the datasets of the three islands in various combinations. Four DTL strategies are illustrated. In the first strategy (DTL Case 1), each of the three DL models is trained using only the Lesvos dataset, while fine-tuning is performed on the dataset of Chios island, in order to create day-ahead predictions for the Chios load. In the second strategy (DTL Case 2), data from both Lesvos and Rhodes concurrently are used for the DL model training period, and fine-tuning is performed on the data from Chios. The third DTL strategy (DTL Case 3) involves the training of the DL models using the Lesvos dataset, and the testing period is performed directly on the Chios dataset without fine-tuning. The fourth strategy is a multi-task deep learning (MTDL) approach, which has been extensively studied in recent years. In MTDL, the three DL models are trained simultaneously on all three datasets and the final predictions are made on the unknown part of the dataset of Chios. The results obtained demonstrate that DTL can be applied with high efficiency for day-ahead load forecasting. Specifically, DTL Case 1 and 2 outperformed MTDL in terms of load prediction accuracy. Regarding the DL models, all three exhibit very high prediction accuracy, especially in the two cases with fine-tuning. The ELM excels compared to the single models. More specifically, for conducting day-ahead predictions, it is concluded that the MLP model presents the best monthly forecasts with MAPE values of 6.24% and 6.01% for the first two cases, the CNN model presents the best monthly forecasts with MAPE values of 5.57% and 5.60%, respectively, and the ELM model achieves the best monthly forecasts with MAPE values of 5.29% and 5.31%, respectively, indicating the very high accuracy it can achieve.