kalomaze: transfer extrapolation should be a learned function, not a theory
kalomaze · x · 2026-09-12
Open-source researcher kalomaze threads on X about why extrapolating transfer is probably something that must be learned: his intuition stems from CoT/reasoning being a decomposition of problems into subroutines that compose generically, so good strategies transfer across very distinct domains.
In the quoted post he argues you don't need a strong mechanistic theory of transfer (which may not exist in closed form) before training — you need a strong learned function of transfer extrapolation for a given policy, which is constructible.
More from AGI Musings
- Critic to AI Safety Crowd: If You Fear Your Tech, Shut It Down Yourself — AIandDesign · 2026-09-12
- AI Doomers Are Actually Utopians, and That's the Road to Dystopia — beffjezos · 2026-09-12
- nc_frey: >10% chance AI ushers in a post-scarcity, disease-free future — inductionheads · 2026-09-12
- RSI is a spectrum, not a switch: from self-hosting C compilers to coding agents — ZeeshanZiaML · 2026-09-12
- AI safety advocates push back on 'regulation is censorship' claim — menhguin · 2026-09-12
- AI solving Navier-Stokes could be the next Deep Blue moment, says Adam Dorr — adam_dorr · 2026-09-12