The Neural Networks of Slow Learning

For the last few days I have been coding up a neural network for my dancing robots AI so that it can learn how to properly decide when to improvise. It isn't really ready yet, but right now it plays at least as good as the current naive approach to the problem (which is to never improvise no matter what.)

I have been training it for a few hours (which should be enough for a simple neural network at least, but it doesn't seem to be catching on very fast (probably I am not doing a good job of coming up with training data)).

I was figuring that the problem was a difficult one (I didn't want to figure it out personally), but I didn't think that it would be this hard. I will continue to fool around with it a bit, but probably not too much. Perhaps the other stupid parts of the AI will fall to neural networks more easily.

Also, I played some more games of Coin Age.

Comments

Post a Comment

Popular posts from this blog

Skye Frontier

Image
I just played a work in progress by Seth Jaffee, so I thought that I would post about it here. The game called Skye Frontier right now. It is an Isle of Skye and The King of Frontier mashup. Since The King of Frontier seems like a Carcassonne/Puerto Rico mashup, this is a second order mashup. Anyway - the game was totally playable, and allowed for strategic decisions to be made. Alison went all out in combining multiple synergistic goals with building a good kingdom. I just went for a degenerate strategy of building the biggest lake possible and then producing and consuming in a cycle. After the game end we summed up our scores using the remaining money tokens, so you can see that I won by 6 points (78-84), so my strategy didn't seem too overpowered, however perhaps it should be nerfed a bit, since it was pretty powerful, and degenerate strategies (in my opinion) should be less powerful than well-thought-out ones. Here is the end state: the reasoning behind my strategy - spelled ou
Read more

More Automatown and getting ready for SaltCon

Image
Alison and I played Automatown with many of the discussed changes last night. We liked some of them, and others we did not. The game ended up ending in 6 rounds, and it took about 80 minutes. The later rounds were longer than the first few rounds. We also were being distracted by kids, so perhaps we could have gone faster. Alison ended up with 7 workers and 9 points while I had 9 workers and 28 points (more about that later). Having fewer types of resources was very good. It made the economy less confusing, and setup and cleanup much faster. The mats worked well, and during the game, we thought of more things to put on them - (the three starting workers can be added here, as well as a reminder of what level is what in terms of dots.) Here is the mat that we played with: We started with 3 scrap, which was not enough given that we were also playing with the rule that made upgrades cost a scrap. In fact, that rule made us strapped for scrap the whole game. We decided that m
Read more

Three more playtests

Image
On Friday I played another game of Skye Frontier with my dad (also Mike), brother in law (Dave), and nephew (Logan). My dad and nephew play games sometimes, but are not super big into the hobby, and my brother in law is moderately a gamer (he buys games and brings them to my game nights, but he doesn't play all of the newest games). Logan started the game (and picked the blue castle), Dave was white, my dad was yellow, and I was red. My dad and I ended up contending over the lakes, while Dave focused on building mountain regions, and Logan built some of everything. I couldn't get a shipping empire set up because boats were hard to come by, and Logan (and Dave) started grabbing cash at an alarming rate, so I instead focused on building up points using scrolls and by closing regions. With a 1 point per sheep scoring tile out there, we all tried to get sheep, and the build for free action was used a lot. I capped off my regions just in the nick of time, and Dave ended the game (wh
Read more