I think the neural part comes from the fact that it use an (hyper) graph of
probabilities like explained by Matt. The symbolic part stems from the fact
that nodes in the graph are labeled which strings which form a programming
language that can take advantage of FOL. My understanding is that
It's a weighted, labeled hypergraph, where a "symbol" can be represented by
a single node or a network-of-nodes/sub-graph. Ideally, Inference and
Hebbian Learning rules allow new "symbols" to form automatically as new
nodes are created to connect groups of related nodes. The weights of the
nodes
Hello All,
i read that, Atomspace is a neural-symbolic network. How?
or is it only a weighted hypergraph? Could any one please explain?
Thanks,
Vishnu
--
You received this message because you are subscribed to the Google Groups
"opencog" group.
To unsubscribe from this group and stop
