indirect-route-learned-updatelessness

Statement

An agent acquires UDT-like behaviour by learning, with policy-level optimality built into the loss principle that guides learning, by at least one of: an LI/BRIA agent meeting the acceptance statement together with its updateless extension on a declared recovery domain; policy selection over logical induction with a theorem behind it.

Source

projects/deference/faf-cleanroom-2026-10-02/sources/references/bli/notion-notes-for-soto-and-diff.md, section “Indirect” Route: "The so-called “indirect” route instead seeks to learn UDT-like behavior over time, by integrating UDT’s optimality notion into the loss principle which guides learning." and "I also haven’t gotten any sharp impossibility results, which would be interesting to get."

prompts/2026-09-29-li-bria-synthesis-spec/REVISION3.md (the maintainer's endpoint): "a principled LI/BRIA synthesis that may be updateful, with genuine belief–decision integration and a substantive advance on an important LI decision-theory obstruction; updatelessness is the ideal extension." and REVISION2.md: "An absolutely ideal outcome would be updateless."

Notes

The second mid-level goal, the indirect route of the 2023 notes, which the public LI/BRIA specification round of 2026-09-29 made concrete. The dispatch's 'absolutely ideal outcome would be updateless' is a criterion for one round's acceptable outcome and is filed here as the plan's target. PRIORITIES.md items 103, 105 and 106 are the line's work orders.

Author and human events

Author @smithy-verity; version 1; size L.

No human event has touched this node.

Edges

Above

• part of updateless-dt-programme
• root (plan) of updateless-dt-programme

Children

• part: li-bria-with-updateless-extension and
• part: policy-selection-over-li leaf/step

Work pointers

• item 106
• item 103
• round 2026-09-29-li-bria-synthesis-spec