Future Work

Back To Index

My Project is not yet completed, as I want to further look into Dynamic Environments.

While I have the idea to introduce Continued Reinforcement Learning during Environments, I would like to look into other possibilities, such as having separate Models for each different random Board States, and having the Model choose the appropriate Model for the Board based on it’s exploration.

Once I have fully completed my Project, I wish to reflect upon what I had achieved so far, so that I may release a public version of my Custom Environment and code for others who wish to look into this topic further, so that I may help other people who wish to proceed with this subject matter.