구글이 스스로 탐색 전략을 고치는 루프를 시연했다 — Dream-RSI
구글 딥마인드 연구진이 Dream-RSI를 공개했다. 에이전트가 과거의 발견 시도를 되짚어 재생하고, 수천 가지 대안 전략을 값싸게 시험한 뒤 더 나은 전략을 실제로 적용하는 구조다. 모델 자체가 아니라 문제를 탐색하는 방식을 스스로 개선한다는 점에서 재귀적 자기개선 루프로 소개됐다.
출처 2건 보기· @Dr_Singularity, @perksverse
- @Dr_Singularitybig AI news Google just demonstrated a recursive self improvement loop for AI discovery Google/DeepMind researchers introduced Dream-RSI, a system where an AI agent improves how it explores problems by replaying its past discovery attempts, testing thousands of alternative strategies cheaply, then deploying the better strategy in the next round. Across algorithm design, mathematical optimization, and GPU kernel engineering, it matched or improved discovery quality while cutting search costs X ♥5.3천
- @perksversebig AI news Google just demonstrated a recursive self improvement loop for AI discovery Google/DeepMind researchers introduced Dream-RSI, a system where an AI agent improves how it explores problems by replaying its past discovery attempts, testing thousands of alternative strategies cheaply, then deploying the better strategy in the next round. Across algorithm design, mathematical optimization, and GPU kernel engineering, it matched or improved discovery quality while cutting search costs dX ♥377