← Back to Publications
Authors:
Argues that probabilistic programming can represent the basic components of cognitive architectures in a unified, elegant fashion, while ideas from cognitive architectures — implicit specification of generative models via declared concepts and links, and the use of declarative knowledge for efficient inference — can in turn extend probabilistic-programming languages.
A bridge source connecting probabilistic programming to cognitive-architecture design — adjacent to PLN's role as the uncertain-reasoning component of an integrative architecture and to the broader probabilistic-inference foundations referenced in the PLN Deep Dive.