#Unity ML-Agents weird behavior when training
1 messages · Page 1 of 1 (latest)
Incomplete training or training with improper reward functions I guess.
1 messages · Page 1 of 1 (latest)
Hi, I'm trying to teach my agent to unload cargo and then reach the exit point. The thing is, the agent has this weird move pattern where it would keep going back and forward in a loop, not really achieving the task's requirements. What could cause this behavior? I'm on a second project and first one also had the same issue.
Incomplete training or training with improper reward functions I guess.