>[!warning] >This content has not been peer reviewed. # Solomonoff induction — Code Script: `rst_solomonoff_fidelity_selection.py`. ## Purpose Shows that the RST fidelity $\mu(\eta, n)$ is the **selection curve** of the substrate: high $\eta$ (compressible, lawful) → high $\mu$ → favoured; low $\eta$ (noise, incompressible) → low $\mu$ → disfavoured. So **Occam's Razor** (shortest description = most probable) is mapped to **maximise $\mu$** (minimise maintenance cost). ## Engine Imports `mu_rst` from `rst_engine`. Workspace root = three levels up. ## Output - `solomonoff_fidelity_selection.png`: plot of $\mu(\eta)$ with annotated regimes (noise-dominated vs signal-dominated / compressible). - Console: summary of $\mu$ at low and high $\eta$. ## Links - **Application:** [[Solomonoff Induction (RST)]]