I made a genetic algorithm that may remedy and optimize Poly Bridge puzzles with synthetic evolution. This Venture’s Supply Code: …
source
Tags: aiartificial intelligenceartificial intelligence newsartificial intelligence news 2023astrosambridgecodingevolvinggeneticlatest news about robotics technologylatest robotslatest robots 2023machine learningPolypoly bridgepolybridgepolybridge 3ProgrammingPuzzlesrobot newsrobotics newsrobotics news 2023robotics technologies llcrobotics technologysam-astrosamastroSolvesunity
While watching Matt(@RealCivilEngineerGaming) tweaking his bridges I thought I'd be cool to replace Matt with genetic algorithm. I don't even thing neural network is required for that.
Basically we can give the bridge (with suboptimal but working design) to the computer and ask it to move joints around until it'd land on the cheapest possible bridge without structural modification. That would drastically speed up the search (compared to generating the bridge from scratch) and hopefully show what the design actually is capable of.
And implementing it as a mod/plugin for the original game would be absolutely cool but i have no idea how hard would it be to integrate such thing into the game.
Astro make your own operating system
Your fitness function could have been based on three things: the changes to the road pieces (max distance change, max angle change) and the max distance change of the balls. Also you should have randomized ball size. With keeping the ball as level as possible the main goal. This will help avoid jumps, hills, swings, and dips. If you’re given a straight path of road joints and told to keep it as level as possibly by only adding bridge connections etc.
a follow up video on this but it minimized stress would be really interesting
-1 for c#
The reward function ends up being as complex as just solving the problem without AI, lol.
I know AI doesn't have emotions, but for this survival of the fittest thing it makes sense they would cheat. Sure, you're not going to win, but you'll live.
So you didnt have the time to do it yourself so you used the time that you DIDNT have to make an ai to do the thing you DIDNT have the time to do
This is like a mix of code bullet and real civil engineer😂😂 you got a new sub
He doesn't have enough time he said.
1:58 instant sub and like for that
Nice code bullet video
reading the comments and just reminded me of the swinging bridge and throw result. a nice simple way to overcome that would be to have 2 balls one after the other test the same bridge like traffic.
"I don't have a lot of time to play this game"
-rebuilds game in unity and builds an AI to play the game which most likely took at least twice as long as playing the game….
9:26 "our reward function was not specific enough, which is actually a big problem with reinforcement learning in…" education systems across the planet lol well said sam
Poop.
4:35 Music?
More polybridge !
We gave something like this as an assignment to students in an evolutionary computation class a while back, and it was a lot of fun. We used a "maximum load" fitness function (keep adding weight until the bridge collapses), and also had students add a cost minimization objective at one point for NSGA-II, though that ran into some fun problems (What bridge minimizes cost? The one where you delete the whole bridge, of course!). On one test problem we were getting really weird results only to find that our scenario was missing one of the two anchor points on the sides of the bridge, so the evolutionary algorithms were desperately trying to keep a huge cantilevered pier from falling over with surprising success!
Architecture, AI, is this a prequel to a Neil Breen movie?
If left to (most) modern engineers we would all live in cube shaped buildings.
I don’t have time to play this game, so I just coded the entire game into an ai platform…
I love how you say genetic algorithms always find a way to cheat the game.
Thats lifes entire story against reality! We keep leveraging ridiculous gimmicks at a biological level to win!
Actually the fact that AI always try to find a way to cheat is creepy
Muito interessante
Nice. Would love to see a more developed version, which takes into account the bridges movements. Less movement = higher scores.
I was hoping to see improving the reward system. The games scoring system is how many vehicles get across and how many pieces are used. This video is about remaking the game more than the ai playing it. There are also some bad assumptions about scoring a bridge.
watching the bridge grow and twitch as it evolved felt like something out of an Alan Resnick film
Wait .. isn’t this how cheat mods are made