Document type

Master thesis

Publication date

Publication license

cc-by-nc-nd (c) Khan, 2025
Please use this identifier to cite or link to this item: https://hdl.handle.net/2445/222550

Observer: An Information-Theoretic Perspective

Journal Title

Journal ISSN

Volume Title

Related resource

Abstract

The boundary between quantum and classical domains remains one of the most profound puzzles in physics, intimately tied to the nature of observation itself. This thesis advances a principled framework wherein observers are recast as System Identification Algorithms (SIAs), finite informational agents1 whose capacity to observe and track external systems is governed by their Kolmogorov complexity. Grinbaum’s hypothesis formalizes observerness2 as an algorithmic resource and gives a relational3 criterion for quantum-classical transitions: a system appears quantum to an observer only when its Kolmogorov complexity lies below that of the observer. Within this framework, classicality emerges as a thermodynamic necessity once memory saturation of the observer forces irreversible erasure, as dictated by Landauer’s principle. We further integrate this perspective into the Local Friendliness experiment, revealing that violations of Local Friendliness inequalities are computationally constrained: they persist only within regimes where complexity gaps between agents remain open. The undecidability of Kolmogorov complexity implies that the precise location of the quantum-classical cut is itself algorithmically inaccessible. We finally interpret the notion of an epistemic horizon discussed in Claim 1 of Restriction A [JM25] through complexity constraints.

Description

Màster Oficial de Ciència i Tecnologia Quàntiques / Quantum Science and Technology, Facultat de Física, Universitat de Barcelona. Curs: 2024-2025. Tutors: Hippolyte Dourdent, Andreas Leitherer.

Citation

Citation

KHAN, Aatif Kaisar. Observer: An Information-Theoretic Perspective. [consulted: 10 of June of 2026]. Available at: https://hdl.handle.net/2445/222550

Export metadata

JSON - METS

Share record