Carl Andersen, P. Basu, Basak Guler, A. Yener, Ebrahim MolavianJazi
{"title":"Protocols for efficient inference communication","authors":"Carl Andersen, P. Basu, Basak Guler, A. Yener, Ebrahim MolavianJazi","doi":"10.1109/PERCOMW.2015.7133993","DOIUrl":null,"url":null,"abstract":"Semantic approaches are increasingly applied to large-scale computing tasks. Nodes in such systems will need to exchange large volumes of declarative information, including ontologies and logical theories. In this paper, we describe our initial work on a system for compressed two-way communication of simplified logical inferences. This new system simplifies protocols for semantic data compression from our prior work, reducing seemingly different operations to hypergraph partitioning. We also present a new protocol that achieves significant speedup, at the cost of reduced compression.","PeriodicalId":180959,"journal":{"name":"2015 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PERCOMW.2015.7133993","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Semantic approaches are increasingly applied to large-scale computing tasks. Nodes in such systems will need to exchange large volumes of declarative information, including ontologies and logical theories. In this paper, we describe our initial work on a system for compressed two-way communication of simplified logical inferences. This new system simplifies protocols for semantic data compression from our prior work, reducing seemingly different operations to hypergraph partitioning. We also present a new protocol that achieves significant speedup, at the cost of reduced compression.