thm-ps-symmetric-mdp-setting
leaf/hypothesis · drafted
Statement
The agent is an optimal policy for a reward function drawn from a distribution invariant under the environmental symmetry, in a finite fully observable Markov decision process.
Source
Turner et al. 2021, section 6, the hypotheses of Theorem 6.13 as stated.
Notes
Definitional within the toy: the theorem is about these agents. The relaxation to trained systems is the three bridges' task, which is why the trained-policy story about the shared assumption lands on them and not here.
Author and human events
Author @smithy-verity; version 1; size S.
No human event has touched this node.
Edges
Above
- part of thm-power-seeking-optimal
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": "19/20",
"range": [
"9/20",
"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.