thm-power-seeking-optimal.record
leaf/record · drafted
Statement
Theorem 6.13 of Turner et al. 2021, average-optimal policies tend to end up in larger sets of recurrent state distributions, is proved in the paper: when a set of recurrent state distributions contains a copy of another via an environmental symmetry and the stated containment conditions hold, most reward functions make it average-optimal to end up in the larger set.
Source
Turner et al. 2021, Theorem 6.13: "Average-optimal policies tend to end up in 'larger' sets of RSDs".
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",
"not in a claims registry"
],
"V": {
"V_prior": null,
"display": "2",
"headline": 2,
"ref": 2,
"valid": 2,
"weaker": "both"
},
"belief": {
"J": 0,
"mid": "91/100",
"range": [
"41/100",
"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.