Reinforcement learning with wordle and open-ai gym
The aim of this project is to build an AI bot that can play the Wordle game, or more generally Squabble
I know there are more deterministic approaches using statistics and log-likelihood that might be more efficient, but I am trying this to for the clickbait title that I’d give my blog if this ends up working
Current status
- Setup wordle-gym environment
- Train some basic algorithms
- Finetune the RL algos to be performant
- Generalize to Squabble esque environment
GitHub