This preliminary note and its sequels present a distributed architecture for AI (Artificial Intelligence) based on a novel market microstructure. The underlying game theory is based on Information-Asymmetric (Signaling) games, where deception is tamed by costly signaling. The signaling, in order to remain honest (e.g., separating), may involve crypto-tokens and distributed ledgers. Here, we will present a rough sketch of the architecture and the protocols it involves. Mathematical and computational analyses will appear in the subsequent sequels.
CITATION STYLE
Mishra, B. (Bud). (2020). Prospero’s Books: A Distributed Architecture for AI: Invited Extended Abstract. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11969 LNCS, pp. 41–49). Springer. https://doi.org/10.1007/978-3-030-36987-3_3
Mendeley helps you to discover research relevant for your work.