search

DylanCope / Multi-Agent-RL-with-TF

Training intrinsically motivated, independent Q-learners to play Tic-Tac-Toe
https://dylancope.github.io/Multiagent-RL-with-TFAgents/
11 stars 5 forks source link

Change ActionSpec to scalar for DQN compatibility #3

Closed npqst closed 3 years ago

npqst commented 3 years ago

Thanks for the great example!

DylanCope commented 3 years ago

Thank you for your contribution, and I'm glad you found the notebook useful :)