This article explores Newcomb's problem through the lens of decision theory, contrasting a "grand-system" view with a "petty-system" perspective. The author argues that decision theories should ideally be compatible with a holistic, grand-system view of the universe, avoiding an artificial separation between the observer and the environment. The piece draws parallels to quantum mechanics, where practical applications often abstract away the observer and measurement devices, focusing on explicit, atom-by-atom models. Similarly, classical decision theory, as exemplified by Savage's work, can be understood through the framework of explicit modeling, where acts map world states to consequences, but the critical issue of the world's state depending on the decider's choice is often overlooked. AI
IMPACT Explores foundational concepts in decision theory relevant to AI alignment and agent design.
RANK_REASON The article is a philosophical exploration of decision theory and Newcomb's problem, drawing parallels to quantum mechanics, rather than reporting on a new release, funding, or event.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →