Spreading Activation over Distributed Microfeatures

One att·empt at explaining human inferencing is that of spreading activat,ion, particularly in the st.ructured connectionist paradigm. This has resulted in t.he building of systems with semantically nameable nodes which perform inferencing by examining t.he pat,t.erns of activation spread. In this paper we demonst.rate t.hat simple structured network infert'ncing can be p(>rformed by passing art.iva.t.ion over the weights learned by a distributed algarit,hm. Thus , an account, is provided which explains a wellbehaved rela t ionship bet.ween structured and distri butt'd conn('ct.ionist. a.pproachrs. INTRODUCTION A primar~· difference brtween t,he nPllral net.works of 20 years ago and t.he ("urrent genera Lion of connect,ionist models is t.he addit.ion of mechanisms whic h permit t.he s),st,em to create all internal represent,ation. These subsymbolic, semantica.lly unnameable, feat.urrs which a.re induced by connectionist. learning algorithms havr been discussed as bt,ing of import. bot,h in structured and distribut.ed ronnl"ctionist nrtworks (cf. Feldman and Balla.rd , 1982; Rumelhart and McClelland, 198(j). The fact that network learning algorit.hms can creatr these rm·cro!eal'ure,s is not, however. enough in itself t.o aC('Qunt for how rognition works. Most. of what, we call int.elligent thought. dt'rives from being able t,o rea:son about. t.he relatioll:::> I)t'tween object.s, to hypothesize about event.s a.nd things, etc. If we are to do cognit.ive modeling we must. complet.e the story by rxplainiJlg how networks can rea,-;oll in tht> wa.y that, humans (or other int,elligrnt beings) do. aile attempt at (-'xplaining such rea.<;oning is that of spreading activat.ion ill the structured cOllllect,ionist. and markerpassing (cf. Charniak, ]983; Hendler, 1987) • The aut.hor is also affiliatf' ci wit.h thp Instit.ut.e for Acivanced C'omputpr Studies a.nd t.he Systems Research Center a.t the Universit.y of Ma.ryland. Funding for this work was provided in part by Offi ce of Naval Resf'arch Grallt N00014-88--K 0,560. 553