Principles of Solomonoff Induction and AIXI

11/25/2011
by   Peter Sunehag, et al.
0

We identify principles characterizing Solomonoff Induction by demands on an agent's external behaviour. Key concepts are rationality, computability, indifference and time consistency. Furthermore, we discuss extensions to the full AI case to derive AIXI.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset