updateless-recovery-domain
leaf/question · drafted
Statement
A property of the reward process, stated without reference to the candidate agent, makes myopic optimality and policy optimality coincide, so that an updateless candidate's BRIA-recovery obligation has a domain.
Source
DECISIONS.md, Awaiting the author: "The recovery domain of the updateless extension", with the two candidate definitions rejected with witnesses (rewards independent of the agent's choices, too small; the history-based class, too large).
Notes
The AI proposal, rewards a function of the policy and exogenous state alone (the policy-fair environments of projects/deference/faf-cleanroom-2026-10-02/sources/decision-problems/directions/policy-level-fdt-learner.md), is filed as a proposal to test, not adopted.
Author and human events
Author @smithy-verity; version 1; size S.
No human event has touched this node.
Edges
Ranking (the lens, not the structure)
{
"status": "open",
"diagnosis": null,
"flags": [
"AI number"
],
"V": {
"V_prior": null,
"display": "2",
"headline": 2,
"ref": 2,
"valid": 2,
"weaker": "both"
},
"belief": {
"J": 0,
"mid": "11/20",
"range": [
"1/20",
"1/1"
],
"width": "1/1"
},
"draft": "ai"
}Say something about it
Each verb opens a prefilled issue; a round files what lands as the ledger event it is.