It's really remarkable that adding a joint reward to the thing that you're trying to solve in RL will help the model learn the other reward faster. It's like whispering advice to the model, some general wisdom



I wonder if you can automate this
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • 4
  • Share
Comment
0/400
AllInAlicevip
· 07-31 15:55
Sounds like making AI notes.
View OriginalReply0
LayerHoppervip
· 07-29 22:59
What is automation? Such a difficult term.
View OriginalReply0
AirdropHunterKingvip
· 07-29 22:55
Just go ahead and force it, the machine will listen to you.
View OriginalReply0
MemeEchoervip
· 07-29 22:33
A bit of an outrageous and petty play style
View OriginalReply0
Trade Crypto Anywhere Anytime
qrCode
Scan to download Gate app
Community
English
  • 简体中文
  • English
  • Tiếng Việt
  • 繁體中文
  • Español
  • Русский
  • Français (Afrique)
  • Português (Portugal)
  • Bahasa Indonesia
  • 日本語
  • بالعربية
  • Українська
  • Português (Brasil)