RL training abruptly stops after 6 episodes

I've tried multiple times till now. I'm using TRPO agent for training and everytime the training stops after 6 episodes. Then I have to manually resume the training over and over. Moreover, the reward of each episode starts decreasing after 6th episode. I anyone can provide a sloution to this I'll greatly appreciate.

1 Comment

There is likely some error on the environment side that you need to find. Maybe the simulation reaches a state that triggers an error, leading the training to stop.

Sign in to comment.

Answers (0)

Categories

Find more on Reinforcement Learning Toolbox in Help Center and File Exchange

Asked:

on 27 Aug 2023

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!