AI Dynamics

Global AI News Aggregator

About

Avoiding self-training bias in agent self-improvement

In SFT we often train on other agents data (distillation) so you’re right. However we also try to climb by self-improvement. This is where it becomes important for the agent not to train on its actions. Any bias on the agent beliefs (weights of the neural net) will be amplified.

→ View original post on X — @nandodf