{"responseHeader":{"status":0,"QTime":6,"params":{"q":"{!q.op=AND}id:\"99668\"","hl":"true","hl.simple.post":"","hl.fragsize":"5000","fq":"!embargo_tdt:[NOW TO *]","hl.fl":"ocr_t","hl.method":"unified","wt":"json","hl.simple.pre":""}},"response":{"numFound":1,"start":0,"docs":[{"file_name_t":"Gu-Parallel_Algorithms.pdf","thumb_s":"/a4/4a/a44af6183a6043521e10f18d2066fcbb3bd7f62e.jpg","oldid_t":"compsci 7781","setname_s":"ir_computersa","restricted_i":0,"format_t":"application/pdf","modified_tdt":"2016-05-25T00:00:00Z","file_s":"/65/44/6544144407ccfe50da2456f49524d41d7676c362.pdf","title_t":"Page 20","ocr_t":"2 tures for solving the CLP problem. These efficient and cost-effective AI computer architectures for constraint-based symbolic processing are capable of offering at least three to ten orders of magnitude performance improvement (i.e., 103 to 1010 times faster) over currently available sequential and parallel CLP algorithms on existing computer machines. In this chapter, some background knowledge for a CLP, including its model, representation, basic computation, and applications are first introduced. Prior art , general procedures and improved techniques for solving a CLP are briefly reviewed. Approaches and philosophy applied in this research are described following a short survey of the current trends in computer architecture design. Major results obtained during this research are summarized. This chapter concludes with an overview of each of the succeeding chapters in this dissertation. 1.1 What 1s a Consistent Labeling Problem (CLP)? The Consistent Labeling Problem [72,73,139] has also been variously referred to as the constraint satisfaction problem [31,40,69,70 ,115,116] and the satisfying assignment problem [52]. A CLP model typically involves a constraint-satisfaction paradigm in which some input data must be given an interpretation that simultaneously satisfies a large set of local constraints. This interpretation corresponds to a pattern of computing activity over the objects, and it is found by an iterative computation in which each object affects many others until the whole system settles down into a stable state. -The consistent labeling problem can be solved by the discrete labeling algorithm or the relaxed model [71 ,83,84,143], Waltz filtering [201 ,202] and the network consistency algorithm [31,115 ,117] .","id":99668,"created_tdt":"2016-05-25T00:00:00Z","parent_i":99969,"_version_":1679953745509613569}]},"highlighting":{"99668":{"ocr_t":[]}}}