Keyboard shortcuts

Press or to navigate between chapters

Press ? to show this help

Press Esc to hide this help

Stable Depth Cardinality Bounds

Abstract

Stable prediction depth is bounded by the available complete-future quotient classes.

Theorem 1.1 (Stable depth bounds for finite runtimes and token carriers).

Proof. Machine-checked in Lean as D5/S3/Observer/Prediction/StableDepthCardinalityBounds.stable_depth_runtime_and_token_bounds (✓ std3). ∎

Source. Repository-derived.

Commentary.

Let Y be a nonempty finite deterministic runtime, let F update its state, and let q map Y surjectively onto the actual output carrier O. The least stable depth is defined by equality of two consecutive finite-future readout relations, while the complete relation compares all future readout coordinates.

The stable finite relation equals the complete-future relation. Therefore the exact finite refinement bound identifies the terminal class count with the cardinality of Y modulo complete-future equality and the initial class count with the cardinality of O.

For a minimal length-L token model, the runtime carrier is the full function type from Fin L to the token alphabet Sigma and the surjective output carrier is Sigma itself. The finite function-cardinality formula then specializes the general bound to |Sigma|^L - |Sigma|.

The source’s final bullets distinguish prediction-classification depth from degradation time, parameter count, cycle-entry time, and semantic memory. Those explanatory contrasts introduce no in-scope predicates and are not asserted as invented universal clauses.

References