Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Genetic Algorithm Walkers

rednuht.org

Post reply on HN
Print view View original

11–20 of 75 posts

Re: Genetic Algorithm Walkers

#11

Post by rednuht »

nhzmju wrote: post #4

I was just about to give up at 70th generation, when the walkers suddenly evolved from 1 to 4 steps in 5 generations. Currently at generation 120 with 7 steps, looks like it is "evolving" in bursts, with long periods of nothing. Well it is stuck at 9 steps at gen 400+. 327 aibobe baeado 948.93 >> 548 aibobe baeado 1058.76 I wonder if the author got any further. I would say this is close to the limit.

It's quite hard to get any better than 10 or so steps, luck starts playing a major role after a while.

Some gait branches seem to reach a dead-end where any improvement would need to come from a major overhaul of the gene combination instead of small mutations.

The terrain starts to get more variation as the distance increases, so that's another piece of evil against the walkers. :D Maybe I should turn that off.

I've let it run for a couple of days. I don't think I got past 12 steps, but it looked pretty regular walking for a while. :)

Re: Genetic Algorithm Walkers

#12

Post by nhzmju »

clemsen wrote: post #10

This looks like fun! It seems that the copied champions are simulated in every run. As long there is no random influence you can probably skip this simulation, as the simulation of the champions is also most likely the most CPU intensive. EDIT: Grammar.

I was thinking the same thing. Since the gene certainly isn't very long the best champions could be stored, until a better champion is found.

Re: Genetic Algorithm Walkers

#13

Post by nhzmju »

rednuht wrote: post #11
nhzmju wrote: post #4

I was just about to give up at 70th generation, when the walkers suddenly evolved from 1 to 4 steps in 5 generations. Currently at generation 120 with 7 steps, looks like it is "evolving" in bursts, with long periods of nothing. Well it is stuck at 9 steps at gen 400+. 327 aibobe baeado 948.93 >> 548 aibobe baeado 1058.76 I wonder if the author got any further. I would say this is close to the limit.

It's quite hard to get any better than 10 or so steps, luck starts playing a major role after a while. Some gait branches seem to reach a dead-end where any improvement would need to come from a major overhaul of the gene combination instead of small mutations. The terrain starts to get more variation as the distance increases, so that's another piece of evil against the walkers. :D Maybe I should turn that off. I've…

I was thinking about writing something similar in C. How long is the genome( bytes/chars )?

Re: Genetic Algorithm Walkers

#14

Post by aselzer »

freehunter wrote: post #5

With a Core i7, I wasn't expecting the simulation to run so slow, even at the lowest settings.. Maybe Firefox on RHEL 6 isn't the optimal platform for this little toy? --edit: it runs much, much faster on Firefox with Windows 8.1. Not a Core i7, but an AMD 8-core thing. Must be Firefox on Linux's implementation, unless it's a video card/driver thing.

Runs pretty fast with Chromium on Linux, but in Firefox it's pretty slow as well. That's with a 2-core 1.7GHz i7

It's probably the physics simulation. Though random JS benchmarks show Chrome and Firefox to be approximately equal in JS performance on my computer, Chrome beating FF slighly on sunspider and the other way around with Octane.

Re: Genetic Algorithm Walkers

#15

Post by Aqwis »

Fun, but even after hundreds of generations the walkers are still pretty bad. How good would they be after thousands or millions of generations? Is it even feasible to create a decent walker using genetic algorithms?

Re: Genetic Algorithm Walkers

#17

Post by mariusz79 »

nhzmju wrote: post #4

I was just about to give up at 70th generation, when the walkers suddenly evolved from 1 to 4 steps in 5 generations. Currently at generation 120 with 7 steps, looks like it is "evolving" in bursts, with long periods of nothing. Well it is stuck at 9 steps at gen 400+. 327 aibobe baeado 948.93 >> 548 aibobe baeado 1058.76 I wonder if the author got any further. I would say this is close to the limit.

Looks like you've hit an evolutionary dead-end.. exterminate!

Re: Genetic Algorithm Walkers

#19

Post by rednuht »

nhzmju wrote: post #13
rednuht wrote: post #11

Earlier quoted context omitted.

It's quite hard to get any better than 10 or so steps, luck starts playing a major role after a while. Some gait branches seem to reach a dead-end where any improvement would need to come from a major overhaul of the gene combination instead of small mutations. The terrain starts to get more variation as the distance increases, so that's another piece of evil against the walkers. :D Maybe I should turn that off. I've…

I was thinking about writing something similar in C. How long is the genome( bytes/chars )?

The genome is pretty simple, just a few motor parameters for each joint in the body.

3 floats for each joint, to be precise. Then those are combined to set the speed of the motor in each joint on every simulation step.

I tried more complex genomes, but it just takes way too long to get any kind of improvement. It didn't make for a fun casual browser experience. :)

Re: Genetic Algorithm Walkers

#20

Post by nhzmju »

mariusz79 wrote: post #17
nhzmju wrote: post #4

I was just about to give up at 70th generation, when the walkers suddenly evolved from 1 to 4 steps in 5 generations. Currently at generation 120 with 7 steps, looks like it is "evolving" in bursts, with long periods of nothing. Well it is stuck at 9 steps at gen 400+. 327 aibobe baeado 948.93 >> 548 aibobe baeado 1058.76 I wonder if the author got any further. I would say this is close to the limit.

Looks like you've hit an evolutionary dead-end.. exterminate!

I felt bad closing the window.
Post reply on HN