A Probabilistic Approach to Marker Propagation

Potential ly, the advantages of marker-passing over local connectionist techniques for associa­ t ive inference are (1) the abi l i ty to differen­ t iate variable bindings, and (2) reduction in the search space and/or number of processing elements. However, the latter advantage has mostly been realized at the expense of accu­ racy and predictabi l i ty. In this paper we con­ sider a class of associative inference to which marker passing is often applied, variously called abductive inference, schema selection, or pat­ tern completion. Analysis of marker seman­ tics in a standard semantic net representation leads to a proposal for more str ict ly regulated marker propagation. An implementat ion strat­ egy employing an augmented relaxation net­ work is out l ined.