‹ BackHN Continuity

Thread

Dream-RSI: Recursive Self-Improvement through Evolving Worlds

213 points · 53 comments · bananaflag

  1. rybosworld · · focus · HN ↗
    Unless I'm misunderstanding, calling this RSI seems misleading?

    This looks like an optimization of current training methods, and a good one, but not "RSI" in the sense of a system that can perpetually improve itself forever.

    1. xidong_wu · · focus · HN ↗
      This paper optimizes a controller/policy which will be used to agent itself in the next round
      1. owenshen24 · · focus · HN ↗
        the use of the phrase here feels very clickbaity tbqh
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.