Filters








11,917 Hits in 8.2 sec

PyTorch Geometric Temporal: Spatiotemporal Signal Processing with Neural Machine Learning Models [article]

Benedek Rozemberczki and Paul Scherer and Yixuan He and George Panagopoulos and Alexander Riedel and Maria Astefanoaei and Oliver Kiss and Ferenc Beres and Guzmán López and Nicolas Collignon and Rik Sarkar
2021 arXiv   pre-print
The main goal of the library is to make temporal geometric deep learning available for researchers and machine learning practitioners in a unified easy-to-use framework.  ...  We present PyTorch Geometric Temporal a deep learning framework combining state-of-the-art machine learning algorithms for neural spatiotemporal signal processing.  ...  A dataset about mass mobility between regions in England and the number of confirmed COVID-19 cases from March to May 2020 [35].  ... 
arXiv:2104.07788v3 fatcat:ktnji6kjrzd7no6blp6zimfutu

Topology dynamics and routing for predictable mobile networks

Daniel Fischer, David Basin, Thomas Engel
2008 Network Protocols (ICNP), Proceedings of the IEEE International Conference on  
We present a model that formalizes predictable dynamic topologies as sequences of static snapshots.  ...  We use this model to design and prove the correctness of a protocol based on link-state routing, whose performance is superior to its static and ad-hoc counterparts.  ...  Part 1: Consider a snapshot transition from S i to S i+1 .  ... 
doi:10.1109/icnp.2008.4697039 dblp:conf/icnp/FischerBE08 fatcat:6ybfq6mqirf67d2spojyqhxpau

Research on the Link Prediction Model of Dynamic Multiplex Social Network Based on Improved Graph Representation Learning (December 2020)

Tianyu Xia, Yijun Gu, Dechun Yin
2020 IEEE Access  
This article studied the dynamic graph representation learning so as to put forward an improved link prediction model in dynamic social network.  ...  INDEX TERMS Link prediction, dynamic social network, graph representation learning. 412 This work is licensed under a Creative Commons Attribution 4.0 License.  ...  ACKNOWLEDGMENT The authors would like to thank the reviewers for their suggestions which helped in improving the quality of the article.  ... 
doi:10.1109/access.2020.3046526 fatcat:truu6tm2lbhp7p55wnuohg7wla

A Blended Deep Learning Approach for Predicting User Intended Actions [article]

Fei Tan, Zhi Wei, Jun He, Xiang Wu, Bo Peng, Haoran Liu, Zhenyu Yan
2018 arXiv   pre-print
It integrates user activity logs, dynamic and static user profiles based on multi-path learning. It exploits historical user records by establishing a decaying multi-snapshot technique.  ...  To overcome these limitations, we propose a novel end-to-end learning scheme to keep track of the evolution of attrition patterns for the predictive modeling.  ...  It is noted that the computation of the user-specific saliency map is very fast due to the requirement of a single back-propagation pass.  ... 
arXiv:1810.04824v1 fatcat:6okce4rlrrd6hn4bpot7zfvmbi

Topological Properties and Temporal Dynamics of Place Networks in Urban Environments

Anastasios Noulas, Blake Shaw, Renaud Lambiotte, Cecilia Mascolo
2015 Proceedings of the 24th International Conference on World Wide Web - WWW '15 Companion  
Understanding the spatial networks formed by the trajectories of mobile users can be beneficial to applications ranging from epidemiology to local search.  ...  We evaluate this model and find it outperforms a number of baseline predictors and supervised learning algorithms on a task of predicting new links in a sample of one hundred popular cities.  ...  Even if a single variable is considered (e.g. distance) and multiple decision boundaries exist, a supervised learning method can learn that more complex relationship.  ... 
doi:10.1145/2740908.2745402 dblp:conf/www/NoulasSLM15 fatcat:7keefvumyzcrfkgwfzqqejfhfa

Topological Properties and Temporal Dynamics of Place Networks in Urban Environments [article]

Anastasios Noulas, Blake Shaw, Renaud Lambiotte, Cecilia Mascolo
2015 arXiv   pre-print
Understanding the spatial networks formed by the trajectories of mobile users can be beneficial to applications ranging from epidemiology to local search.  ...  We evaluate this model and find it outperforms a number of baseline predictors and supervised learning algorithms on a task of predicting new links in a sample of one hundred popular cities.  ...  Even if a single variable is considered (e.g. distance) and multiple decision boundaries exist, a supervised learning method can learn that more complex relationship.  ... 
arXiv:1502.07979v2 fatcat:g5txj3y5zveszcybmjxvlqwnj4

The initial development of object knowledge by a learning robot

Joseph Modayil, Benjamin Kuipers
2008 Robotics and Autonomous Systems  
We describe how a robot can develop knowledge of the objects in its environment directly from unsupervised sensorimotor experience.  ...  We evaluate how well this intrinsically acquired object knowledge can be used to solve externally specified tasks including object recognition and achieving goals that require both planning and continuous  ...  Research of the Intelligent Robotics lab is supported in part by grants from the National Science Foundation (IIS-0413257 and IIS-0538927), from the National Institutes of Health (EY016089), and by an  ... 
doi:10.1016/j.robot.2008.08.004 pmid:19953188 pmcid:PMC2603070 fatcat:j5cdmk2dinbbvp6ukg63fv7jh4

Learning Hierarchical Object Maps Of Non-Stationary Environments with mobile robots [article]

Dragomir Anguelov, Rahul Biswas, Daphne Koller, Benson Limketkai, Sebastian Thrun
2012 arXiv   pre-print
Building models, or maps, of robot environments is a highly active research area; however, most existing techniques construct unstructured maps and assume static environments.  ...  Additionally, we develop a Bayesian model selection algorithm that enables the robot to estimate the total number of objects and object templates in the environment.  ...  Nevertheless, we believe that this work is unique in its ability to learn hierarchical object models in mobile robotics.  ... 
arXiv:1301.0551v1 fatcat:qxr2x6cfcrfzvlhftpejtozbcm

Feature Subsumption for Sentiment Classification of Dynamic Data in Social Networks using SCDDF

Jayanag. B, Vineela. K, Dr. Vasavi.
2012 International Journal of Advanced Computer Science and Applications  
But on a given concept opinions may vary from site to site. Also the past works did not consider the opinions at aggregate level.  ...  Our method takes as input a collection of comments from the social networks and outputs ranks to the comments within each site and finally classifies all comments irrespective of the site it belongs to  ...  RELATED WORK Previous works concentrated on opinions in individual sites and also limited the data set to a single line comment or static data or a limit on number of characters.  ... 
doi:10.14569/ijacsa.2012.030906 fatcat:kf73bvnrofg4zfowj3e3jwafim

Understanding and measuring the urban pervasive infrastructure

Vassilis Kostakos, Tom Nicolai, Eiko Yoneki, Eamonn O'Neill, Holger Kenn, Jon Crowcroft
2008 Personal and Ubiquitous Computing  
The increasing popularity of mobile computing devices has allowed for new research and application areas.  ...  We focus on describing the metrics for describing this infrastructure and elaborate on a set of observation, analysis and simulation methods for capturing, deriving and utilising those metrics.  ...  The stationary devices are connected to a central server aggregating the data in a single database. We provide parts of this infrastructure to other researchers under the GPL license 11 .  ... 
doi:10.1007/s00779-008-0196-1 fatcat:tpjo5qibqvejjnyyxdkg45422u

Instrumenting the City: Developing Methods for Observing and Understanding the Digital Cityscape [chapter]

Eamonn O'Neill, Vassilis Kostakos, Tim Kindberg, Ava Fatah gen. Schiek, Alan Penn, Danaë Stanton Fraser, Tim Jones
2006 Lecture Notes in Computer Science  
Here we describe how we have combined scanning for discoverable Bluetooth devices with two such methods, gatecounts and static snapshots.  ...  To understand the city as a system encompassing physical and digital forms and their relationships with people's behaviours, we are developing, applying and refining methods of observing, recording, modelling  ...  Extending Static Snapshots to Include Bluetooth Interaction Spaces We also extended the static snapshot method with Bluetooth scanning, drawing on the lessons learned in developing and refining our augmented  ... 
doi:10.1007/11853565_19 fatcat:rqpr4kdfafbdbdii6c3sxh4pwi

Compact internal representation of dynamic situations: neural network implementing the causality principle

José Antonio Villacorta-Atienza, Manuel G. Velarde, Valeri A. Makarov
2010 Biological cybernetics  
Then the diffusion-like process relaxes the remaining neurons to a stable steady state, i.e., a CIR is given by a single point in the multidimensional phase space.  ...  First, a wavefront representing the agent's virtual present interacts with mobile and immobile obstacles forming static effective obstacles in the network space.  ...  This study has been sponsored by the EU grant SPARK II (FP7-ICT-216227) and by the Spanish Ministry of Education and Science under the grant FIS2007-65173 and a Juan de la Cierva fellowship (J.A.V.).  ... 
doi:10.1007/s00422-010-0398-2 pmid:20589508 fatcat:recn63734bbmlafrf46oi4mtom

Predictable Mobile Routing for Spacecraft Networks

Daniel Fischer, David Basin, Knut Eckstein, Thomas Engel
2013 IEEE Transactions on Mobile Computing  
We use this model to design and evaluate a predictable mobile-routing protocol based on link-state routing, whose performance is superior to its static and ad-hoc counterparts.  ...  We have developed a model that formalizes predictable dynamic topologies as sequences of static snapshots.  ...  Approach: We present a formal topology model for predictable mobile networks that describes the topology evolution by a sequence of static network-topology snapshots.  ... 
doi:10.1109/tmc.2012.93 fatcat:im3ulhspb5fz3klhbdmj34zgy4

A Survey on Dynamic Network Embedding [article]

Yu Xie, Chunyi Li, Bin Yu, Chen Zhang, Zhouhua Tang
2020 arXiv   pre-print
and the temporal dynamics, which is beneficial to multifarious downstream machine learning tasks.  ...  Compared to widely proposed static network embedding methods, dynamic network embedding endeavors to encode nodes as low-dimensional dense representations that effectively preserve the network structures  ...  Acknowledgement The authors wish to thank the editors and anonymous reviewers for their valuable comments and helpful suggestions which greatly improved the paper's quality.  ... 
arXiv:2006.08093v1 fatcat:3t7ma6zp4rhy5csilzbuwm6k7y

Passenger Mobility Prediction via Representation Learning for Dynamic Directed and Weighted Graph [article]

Yuandong Wang and Hongzhi Yin and Tong Chen and Chunyang Liu and Ben Wang and Tianyu Wo and Jie Xu
2021 arXiv   pre-print
As both spatial and temporal patterns are indispensable passenger demand prediction, relevant research has evolved from pure time series to graph-structured data for modeling historical passenger demand  ...  However, existing graph-based solutions fail to simultaneously consider those three crucial aspects of dynamic, directed, and weighted (DDW) graphs, leading to limited expressiveness when learning graph  ...  Visualizing Learned Passenger Demand and Mobility Patterns (RQ6) In order to better understand the latent patterns learned by Gallat, we use Figure 8 to visualize a part of passengers' demand patterns  ... 
arXiv:2101.00752v1 fatcat:ja4jm3mbrzcu5ho7dwlysmcbvy
« Previous Showing results 1 — 15 out of 11,917 results