OpenCog NS: A Deeply-Interactive Hybrid Neural-Symbolic Cognitive Architecture
Draft — This content has not been approved for publication.
← Back to Publications
OpenCog NS: A Deeply-Interactive Hybrid Neural-Symbolic Cognitive Architecture Designed for Global/Local Memory Synergy
Authors:
Year: 2009
Venue: Artificial General Intelligence (AGI 2009)
Links: paper (PDF)
Summary
Defines a “deeply-interactive” hybrid neural-symbolic architecture in which neural-net and symbolic components interact frequently enough to form a combined dynamical system, and describes OpenCog NS — OpenCog integrated with a hierarchical attractor neural network (HANN). In it, symbolic reasoning intervenes in attractor formation within the HANN while the HANN in turn helps guide logical inference and evolutionary program learning.
Relevance to Hyperon
An early neural-symbolic integration source in the OpenCog/PRIMUS lineage — relevant to the PRIMUS Deep Dive's account of a deeply-integrated cognitive architecture combining symbolic, evolutionary, and connectionist components.