As a deep learning model typically contains millions of trainable weights, there has been a growing demand for a more efficient network structure with reduced storage space and improved run-time efficiency. Pruning is one of the most popular network compression techniques. In this paper, we propose a novel unstructured pruning pipeline, Attention-based Simultaneous sparse structure and Weight Learning (ASWL). In ASWL, an efficient algorithm is proposed to calculate the pruning ratios layer-wisely from attentions, and both weights for the dense network and the sparse network are tracked so that the pruned structure is simultaneously learned from randomly initialized weights. Our experiments on MNIST, Cifar10, and ImageNet show that ASWL achieves superior pruning results in terms of accuracy, pruning ratio and operating efficiency when compared with state-of-the-art network pruning methods.

We apply reinforcement learning to video compressive sensing to adapt the compression ratio. Specifically, video snapshot compressive imaging (SCI), which captures high-speed video using a low-speed camera is considered in this work, in which multiple ( B ) video frames can be reconstructed from a snapshot measurement. One research gap in previous studies is how to adapt B in the video SCI system for different scenes. In this article, we fill this gap utilizing reinforcement learning (RL). An RL model, as well as various convolutional neural networks for reconstruction, are learned to achieve adaptive sensing of video SCI systems. Furthermore, the performance of an object detection network using directly the video SCI measurements without reconstruction is also used to perform RL-based adaptive video compressive sensing. Our proposed adaptive SCI method can thus be implemented in low cost and real time. Our work takes the technology one step further towards real applications of video SCI.

As the SSD technology develops, quad-level cell (QLC) NAND based SSD is gradually being introduced to the market. And as such, we evaluate the QLC technologys impact on the landscape of modern datacenters. Since a large number of applications and workloads in the modern datacenters have far more read requests than they write requests, QLC SSD provides a promising solution. For example, real-time analytics and big data, machine and deep learning, and read-intensive AI applications are all read hungry perfectly suited for the QLC. Its favorable performance (especially in reads), high capacity and density greatly help modern datacenters to provide more efficient services to their customers. At the same time, the low cost of QLC SSD also helps to lower the cost of operation for datacenters. Additionally, we explore the state-of-art QLC SSD from the system architecture point of view to shows its key advancements from previous technologies. By conducting a comprehensive performance evaluation of QLC SSD, we are able to compare it with other types of SSD and analyze factors that impact its performance.

In recent years, NAND flash-based solid state drives (SSD) have been widely used in datacenters due to their better performance compared with the traditional hard disk drives. However, little is known about the reliability characteristics of SSDs in production systems. Existing works study the statistical distributions of SSD failures in the field. However, they do not go deep into SSD drives and investigate the unique error types and health dynamics that distinguish SSDs from hard disk drives. In this paper, we explore the SSD-specific SMART (Self-Monitoring, Analysis, and Reporting Technology) attributes to conduct an in-depth analysis of SSD reliability in a production environment. Data is collected from a scalable production system having several physical locations. Our dataset contains over a million records with more than twenty attributes. We leverage machine learning technologies, specifically data clustering and correlation analysis methods, to discover groups of SSDs which have different health status and relations among SSD-specific SMART attributes. Our results show that 1) Media wear affects the reliability of SSDs more than any other factors, and 2) SSDs transit from one health group to another which infers the reliability degradation of those drives. To the best of our knowledge, this is the first study that investigates SSD-specific SMART data to characterize SSD reliability in a production environment.

Data protection is the process of backing up data in case of a data loss event. It is one of the most critical routine activities for every organization. Detecting abnormal backup jobs is important to prevent data protection failures and ensure the service quality. Given the large scale backup endpoints and the variety of backup jobs, from a backup-as-a-service provider viewpoint, we need a scalable and flexible outlier detection method that can model a huge number of objects and well capture their diverse patterns. In this paper, we introduce H2O, a novel hybrid and hierarchical method to detect outliers from millions of backup jobs for large scale data protection. Our method automatically selects an ensemble of outlier detection models for each multivariate time series composed by the backup metrics collected for each backup endpoint by learning their exhibited characteristics. Interactions among multiple variables are considered to better detect true outliers and reduce false positives. In particular, a new seasonal-trend decomposition based outlier detection method is developed, considering the interactions among variables in the form of common trends, which is robust to the presence of outliers in the training data. The model selection process is hierarchical, following a global to local fashion. The final outlier is determined through an ensemble learning by multiple models. Built on top of Apache Spark, H2O has been deployed to detect outliers in a large and complex data protection environment with more than 600,000 backup endpoints and 3 million daily backup jobs. To the best of our knowledge, this is the first work that selects and constructs large scale outlier detection models for multivariate time series on Big Data platforms.

Bilingual sentiment lexicon is fundamental resource for cross-language sentiment analysis but its compilation remains a major bottleneck in computational linguistics. Traditional word alignment algorithm faces with the status of large alignment space, which may introduce redundant computations as well as alignment errors. In this paper, we use collocation alignment to extract bilingual sentiment lexicon overcoming the drawbacks of word alignment. The idea of collocation alignment is inspired by the strong cohesion between feature words and opinion words in sentiment corpus. Experimental results show that our approach not only decreases the computing time dramatically but also improves the precision of extracted bilingual word pairs due to the smaller alignment space.