tag:blogger.com,1999:blog-1180536024131440638.post3085589101202134355..comments2023-10-02T02:12:54.512+11:00Comments on Project AGI: Some interesting finds: Acyclic hierarchical modelling and sequence unfoldingUnknownnoreply@blogger.comBlogger2125tag:blogger.com,1999:blog-1180536024131440638.post-10732000964327410052016-04-25T12:27:15.701+10:002016-04-25T12:27:15.701+10:00Looking at the Felleman & Essen ref:
".....Looking at the Felleman & Essen ref:<br /><br />"...the pure form of this hypothesis is difficult to reconcile with the finding of highly reciprocal connectivity and parallel channels discovered in more recent studies of the visual pathway (cf. Rockland and Pandya, 1979; Stone et al., 1979; Lennie, 1980; Lennie et al., 1990; Shapley, 1990). On the other hand, there is no a priori reason to restrict the notion of hierarchical processing to a strictly serial sequence."<br /><br />It sounds like they are breaking a sweat trying to account for an avalanche of evidence for feedback.<br /><br />Why are you guys against feedback anyway?<br /><br />I feel the same about Bengio's paper. Feels like he's working hard to squeeze the possibility of his model out of the evidence, despite it not fitting naturally.<br /><br />Like he's saying "we could still get this round peg into this square hole, they don't absolutely exclude each other."rob_freemanhttps://www.blogger.com/profile/00126505256265314963noreply@blogger.comtag:blogger.com,1999:blog-1180536024131440638.post-13932766476554981092016-04-25T12:07:43.920+10:002016-04-25T12:07:43.920+10:00I looked up the papers referred to by Bengio for t...I looked up the papers referred to by Bengio for the "acyclic" claim. On the way I found this paper:<br /><br />http://www.ini.uzh.ch/admin/extras/doc_get.php?id=42092<br /> Canonical Cortical Circuits<br /> Rodney J. Douglas and Kevan A. C. Martin <br /><br />'More recent experimental and theoretical considerations of the cortical circuits, however,have suggested a rather different architecture: one in which local circuits of cortical neurons are connected in a series of nested positive and negative feedback loops, called “recurrent circuits”'<br /><br />I'm guessing you mean "feedback" up and down layers? But I'd like to put my finger on exactly what this "acyclic" claim refers to.rob_freemanhttps://www.blogger.com/profile/00126505256265314963noreply@blogger.com