Will any agent perform better on Minecraft (or comparable open world game) after being fine-tuned on a manual by 2027?
One-OffIA2mManifold MarketsSenza KYCDati di risoluzione verificatiBen calibratoPrevisione della comunità attualeSì 79.8%
Previsori
17
Tipo di domanda
binary
Metodologia
Play-money forecasting platform
Tipo di fonte
Previsione
Dati di mercato
Aggiornato 17 ore fa
Obsoleto11 lug 22, 19:56 UTC1 gen 27, 05:00 UTC
Trend
Regole
To clarify: the experiment is that there are two copies of an agent that runs on Minecraft (or some other open world game environment).
- The agent has the capacity to be fine-tuned with text.
- One version is passed a manual for the game as text (or text + images, but *not* video), the other runs without any finetuning.
- Will the former perform better than the latter (either better sample efficiency or better final reward)?
- The agent can't have been trained on that env before, but it can be trained on other envs/data beforehand (e.g. it's okay if there's a pretrained LLM in the loop).
