Lemma 3.2 — Evaluation at mean parameters
ProvedXuDRMDP.Finite.lemma_3_2Let be a finite nonempty staged state set with finite nonempty action sets , initial stage consisting of alone, finite horizon , zero terminal rewards and discount . Assume the nested local parameter sets () are nonempty, convex and compact sets of valid transition/reward vectors. Local sets may couple all actions. Let . The local class consists of all Borel probability laws supported on with , and consists of their statewise product laws. Policies are all history-dependent randomized policies on state/action histories, including null histories. Returns are the independent finite trajectory expectations, subsequently integrated against the parameter law.
For every policy and every , let . Then
The left side integrates the independently defined trajectory return. The right side evaluates that same path sum at the mean parameter vector. No optimality, Markov property, or deterministic-action restriction is imposed on .
Differences from the source:
- encoding: finite states and state-dependent actions, one-based staged histories, and expected discounted trajectory returns → nonempty Fin types, zero-based stages, total policy extensions on unused histories, and explicit finite trajectory sums followed by Bochner integration; reason: finite relabeling and unused extensions preserve all source instances; compact outer support ensures integrability and finite bounded returns
import Definitions.Def_XuDRMDP_Finite_Setup import Definitions.Def_XuDRMDP_Finite_Policies import Definitions.Def_XuDRMDP_Finite_Ambiguity import Definitions.Def_XuDRMDP_Finite_Returns import Definitions.Def_XuDRMDP_Finite_Robustness open MeasureTheory
namespace XuDRMDP.Finite
theorem lemma_3_2 {n : ℕ} [NeZero n] {a : Fin n → ℕ} [∀ s, NeZero (a s)]
(G : Setup n a) (U : Uncertainty G) (π : Policy a)
(ν : Measure (Parameters n a)) (hν : ν ∈ globalClass U) :
performance G π ν = totalReturn G π (globalMean ν) := by sorry
end XuDRMDP.Finite
Read-back
What the Lean code literally says, in plain math · claude-opus-5-5
Data.
- states and actions with every .
- A setup : horizon , discount , stage map , and initial state , the unique state of stage .
- An uncertainty structure : confidence levels , and nested nonempty convex compact sets of valid local MDP coordinates in . Valid means the point equals some imported-type MDP's data at , has zero rewards if , and otherwise zero transition entries to states not at stage .
- An arbitrary policy , assigning to every history a probability vector on the current state's actions.
- A measure on in the global class, i.e. for some family with each a probability measure, and for all .
Total return. For a parameter , the total return is the finite sum, over sequences () and final states , of
where:
- ;
- is the history with past and current state .
Mean parameter. is the parameter with , a vector-valued Bochner integral.
Claim.
Degenerate cases.
- If , both sides are : the sum runs over the single empty sequence with zero reward.
- Integrals are Bochner integrals, equal by convention to (or the zero vector) for non-integrable integrands. The identity is asserted with that convention.
- is not required to lie in any or to be valid.
- The transition factors are raw real coordinates with no normalisation required by the statement.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.