作者:
Luo, XinZhou, MengChuDongguan Univ Technol
Sch Comp Sci & Technol Dongguan 523808 Peoples R China Chinese Acad Sci
Chongqing Engn Res Ctr Big Data Applicat Smart Ci Chongqing 400714 Peoples R China Chinese Acad Sci
Chongqing Inst Green & Intelligent Technol Chongqing Key Lab Big Data & Intelligent Comp Chongqing 400714 Peoples R China New Jersey Inst Technol
Dept Elect & Comp Engn Newark NJ 07102 USA
High-dimensional and sparse (HiDS) matrices from recommender systems contain various useful patterns. A latent factor (LF) analysis is highly efficient in grasping these patterns. Stochastic gradient descent (SGD) is ...
详细信息
High-dimensional and sparse (HiDS) matrices from recommender systems contain various useful patterns. A latent factor (LF) analysis is highly efficient in grasping these patterns. Stochastic gradient descent (SGD) is a widely adopted algorithm to train an LF model. Can its extensions be capable of further improving an LF models' convergence rate and prediction accuracy for missing data? To answer this question, this work selects two of representative extended SGD algorithms to propose two novel LF models. Experimental results from two HiDS matrices generated by real recommender systems show that compared standard SGD, extended SGD algorithms enable an LF model to achieve a higher prediction accuracy for missing data of an HiDS matrix, a faster convergence rate, and a larger model diversity.
Industrie 4.0 and data analytics blur the separation of operational and information technology that prevailed for industrial automation over the last decades. Decentralized control systems for production plants and ro...
详细信息
Industrie 4.0 and data analytics blur the separation of operational and information technology that prevailed for industrial automation over the last decades. Decentralized control systems for production plants and robot cells collaborate actively with higher-level systems for bigdata analytics. In parallel, the complexity of designing and operating a system architecture for data collection and analysis increases dramatically as more experts from different domains get involved. Graphical modeling notations facilitate the design process by formalizing implicit knowledge, but currently do not exist for the combined description of field layer and data analytics. Modeling the system architecture, relevant constraints, and requirements early during the design process can increase the efficiency of the system development and deployment, especially as experts with various backgrounds are involved. In this contribution, a new graphical notation is introduced and evaluated in three industrial case-studies. The notation describes the underlying hardware and software components of cyber-physical systems of systems, the flow of data, and relevant constraints. The evaluation proved that the notation is powerful in supporting the engineering of data collection and analysis architectures in industrial automation. Future work is related to extending the scope of the modeling approach to include safety applications and real-time considerations on the field level.
Developing an automated vehicle, that can handle complicated driving scenarios and appropriately interact with other road users, requires the ability to semantically learn and understand driving environment, oftentime...
详细信息
Developing an automated vehicle, that can handle complicated driving scenarios and appropriately interact with other road users, requires the ability to semantically learn and understand driving environment, oftentimes, based on analyzing massive amounts of naturalistic driving data. An important paradigm that allows automated vehicles to both learn from human drivers and gain insights is understanding the principal compositions of the entire traffic, termed as traffic primitives. However, the exploding data growth presents a great challenge in extracting primitives from high-dimensional time-series traffic data with various types of road users engaged. Therefore, automatically extracting primitives is becoming one of the cost-efficient ways to help autonomous vehicles understand and predict the complex traffic scenarios. In addition, the extracted primitives from raw data should 1) be appropriate for automated driving applications and also 2) be easily used to generate new traffic scenarios. However, existing literature does not provide a method to automatically learn these primitives from large-scale traffic data. The contribution of this letter has two manifolds. The first one is that we proposed a new framework to generate new traffic scenarios from a handful of limited traffic data. The second one is that, we introduce a nonparametric Bayesian learning method-a sticky hierarchical Dirichlet process hidden Markov model-to automatically extract primitives from multidimensional traffic datawithout prior knowledge of the primitive settings. The developed method is then validated using one day of naturalistic driving data. Experiment results show that the nonparametric Bayesian learning method is able to extract primitives from traffic scenarios where both the binary and continuous events coexist.
Self-driving vehicles rely on detailed semanticmaps of the environment for operating. In this letter, we propose a method to autonomously generate such a semantic map enriched with knowledge of parking spot locations....
详细信息
Self-driving vehicles rely on detailed semanticmaps of the environment for operating. In this letter, we propose a method to autonomously generate such a semantic map enriched with knowledge of parking spot locations. Our method detects and uses parked vehicles in the surroundings to estimate parking lot topology and infer vacant parking spots via a graph-based approach. We show that our method works for parking lot structures in different environments, such as structured parking lots, unstructured/unmarked parking lots, and typical suburban environments. Using the proposed graph-based approach to infer the parking lot structure, we can extend the estimated parking spots by 57%, averaged over six different areas with ten trials each. We also show that the accuracy of our algorithm increases when combining multiple trials over multiple days. With ten trials combined, we managed to estimate the whole parking lot structure and detected all parking spots in four out of the six evaluated areas.
Dropout strategy is a simple and common regularization method in the construction of deep network that it can control the status of units in the Dropout layers according to the constant probability values in the train...
详细信息
ISBN:
(纸本)9781728103778
Dropout strategy is a simple and common regularization method in the construction of deep network that it can control the status of units in the Dropout layers according to the constant probability values in the training processes to prevent the training from overfitting. However, the probability values of the Dropout strategy are single and decided by users, which means that we need more training iterations to receive better results and avoid less fitting problem. In this paper, two evolutionary algorithms, genetic algorithm and differential evolution algorithm are used to optimize the set probability values of network units to improve dropout strategy and they are proved to be able to increase the accuracy of the original method to about 5%.
To enable robots to work on real home environments, they have to not only consider common knowledge in the global society, but also be aware of existing rules there. Since such "local rules" are not describa...
详细信息
ISBN:
(纸本)9781538680940
To enable robots to work on real home environments, they have to not only consider common knowledge in the global society, but also be aware of existing rules there. Since such "local rules" are not describable beforehand, robot agents must acquire them through their lives after deployment. To achieve this, we developed a framework that a) lets robots record long-term episodic memories in their deployed environments, b) autonomously builds probabilistic object localization map as structurization of logged data and c) make adapted task plans based on the map. We equipped our framework on PR2 and Fetch robots operating and recording episodic memory for 41 days with semantic common knowledge of the environment. We also conducted demonstrations in which a PR2 robot tidied up a room, showing that the robot agent can successfully plan and execute local-rule-aware home assistive tasks by using our proposed framework.
Inmany 3-D perception applications, ground segmentation is a necessary preprocessing phase together with point cloud cleaning and outlier removal. This letter presents a method for ground segmentation in large-scale p...
详细信息
Inmany 3-D perception applications, ground segmentation is a necessary preprocessing phase together with point cloud cleaning and outlier removal. This letter presents a method for ground segmentation in large-scale point clouds of industrial environments acquired using a terrestrial laser scanner (TLS). TLSs provide high-precision, dense 3-D measurements, and therefore, such instruments are becoming the state of the art technology for surveying tasks. In contrast to many previousworks, where ground segmentation has been investigated using a single scan (e.g., in LiDAR-equipped vehicles), experiments have been performed in large-scale point clouds that contain over 1010 points measured from multiple scan stations. The proposed solution is based on a robust estimation of points belonging to the ground below each scan station and it can be applied even in challenging scenarios with nonplanar regions.
暂无评论