The science behind Senua AI.

Senua AI did not invent a new theory of intelligence. It applies one with more than three decades of peer-reviewed work behind it: computational mechanics, the theory of causal states and epsilon-machines, developed by James P. Crutchfield and his colleagues.

"Do not distinguish histories that lead to the same predictions of the future."

James P. Crutchfield

Crutchfield is Distinguished Professor of Physics at the University of California, Davis, where he directs the Complexity Sciences Center, and external faculty at the Santa Fe Institute. In 1980, with Packard, Farmer and Shaw, he showed how to reconstruct a system's state from a series of its own measurements. In 1989, with Karl Young, he introduced causal states and the epsilon-machine, and named the field computational mechanics: a theory of how a process stores and transforms information, and so of its structure.

That theory answers the question every learning system faces: which parts of the past matter for the future? Senua AI is built on its answer.

Senua AI builds on this published work. Citing it here does not imply that its authors endorse Senua AI.