Esoteric Update #161 - Training The Hand That Pets

Once more, no lore this time, but I have an assortment of things I hope to be interesting.

A Primer On The Aeons
Mechanics: Anticipating The Player With Q-Learning
Public Game Design Document: 29.11.21

In terms of the update... well... primarily, it concerns the second link, and there's little more I can add that's already detailed in there. But, for the most part, I've spent the week learning about Q-Leaning, implementing it in C#, then testing it, re-implementing it in Esoteric Script, re-testing it... then building a visual wrapper for it.

Now, there's one thing I need to point to. While the FMDPs I've worked on so far have been relatively small, I can see this still being useful. And some of the tests I've done have proven to detect sequences and patterns.

I've been looking into maybe Lambda-Q-Learning to be a possible extension to what I have right now, but I think I'll hold off on that. We're preparing some new clothing options, and I'll be going back to writing now. Improving the implementation can wait.