Open jstac opened 8 years ago
Saw that awhile back, but code is GPL'd. Looking at or using it as the basis of our code would require us to have a GPL license as well, so unless we can get it licensed differently (under MIT or BSD for example) then, in my opinion, it is off the table.
I've been meaning to reach out to @babrooks about it, but haven't actually sat down to write the email. He may have his hands tied in terms of licensing because of GPL dependencies. The original paper (and a follow up that he is working on) are both very nice and I think are likely the best way to solve for the value sets of two player games.
The only issues so far with actually implementing the algorithm are a lack of computational geometric tools in Julia (intersecting some sets). I had started building some of the pieces that I needed to implement it, but stopped because of time constraints -- I wanted to use good computational geometry algorithms to make it as slick as possible, but, in retrospect, just getting it working would be a nice first pass (I tried not to take advantage of some of the structure that the problem gave). To get a working version of these algorithms we could talk to @squipbar. He has implemented a lot of these pieces in his Polygons.jl package.
Just noting the existence of this very nice code for dynamic games:
http://www.benjaminbrooks.net/software.shtml
cc @babrooks cc @cc7768
Anyone interested in building Julia versions...?