Code center > Suggestions
GPGPU acceleration?
Numsgil:
--- Quote from: Billy on April 28, 2017, 09:30:25 AM ---I guess you're probably right. I was thinking that if 90% of the bots have very similar DNA and are doing exactly the same activity, there would be very little divergence and it would effectively be SIMD. I'd still like to make a prototype, maybe with some minimal language rather than DNA, so I can compare the performance with that on a CPU.
--- End quote ---
I don't think it's impossible with a bit of ingenuity to come up with something, but it's a bit awkward. At their core, bots have 1000 inputs (their memory) and 1000 outputs (their memory after storing stuff in to it). GPUs don't really like mapping inputs to outputs in the large scale like that. They tend to be more comfortable mapping a small fixed number of inputs to 1 output over and over in parallel, and doing multiple passes if they need to.
One idea: I could imagine a DNA language built around the idea of smaller DNA programs feeding to each other. Like, imagine a neural network, but instead of a sum and logistic function each node is a small program. If each small program had 16 inputs, say, and a single output that fed on to the next layer of nodes, each node could be executed massively in parallel using the GPUs quite nicely. I don't know if this would be even remotely efficient, but it would sort of map to how the hardware works at least.
Billy:
--- Quote from: Numsgil on April 28, 2017, 08:46:07 PM ---
--- Quote from: Billy on April 28, 2017, 09:30:25 AM ---I guess you're probably right. I was thinking that if 90% of the bots have very similar DNA and are doing exactly the same activity, there would be very little divergence and it would effectively be SIMD. I'd still like to make a prototype, maybe with some minimal language rather than DNA, so I can compare the performance with that on a CPU.
--- End quote ---
I don't think it's impossible with a bit of ingenuity to come up with something, but it's a bit awkward. At their core, bots have 1000 inputs (their memory) and 1000 outputs (their memory after storing stuff in to it). GPUs don't really like mapping inputs to outputs in the large scale like that. They tend to be more comfortable mapping a small fixed number of inputs to 1 output over and over in parallel, and doing multiple passes if they need to.
One idea: I could imagine a DNA language built around the idea of smaller DNA programs feeding to each other. Like, imagine a neural network, but instead of a sum and logistic function each node is a small program. If each small program had 16 inputs, say, and a single output that fed on to the next layer of nodes, each node could be executed massively in parallel using the GPUs quite nicely. I don't know if this would be even remotely efficient, but it would sort of map to how the hardware works at least.
--- End quote ---
So there would be a fixed set of these node functions that, arranged in different tree structures, produce different bot behaviours? With a pre-written kernel for each function?
Billy:
--- Quote from: Botsareus on April 28, 2017, 04:51:08 PM ---I get I am not good with words; it is not my native.
However point still stands that no one figured it out. It would have been perfect architecture because each master process can just assign threads or processes or whatever to different sized CPUs. Or just hack it into a programming language. Something like "start a new n speed thread"
--- End quote ---
Not sure it's a case of figuring out how (though I don't know much about electronics), rather proving that it's worth doing. Most things can be done well enough on a CPU and/or GPU. Maybe having a few smaller cores is no better than having a single extra full-size core.
Numsgil:
--- Quote from: Billy on April 29, 2017, 08:20:50 AM ---So there would be a fixed set of these node functions that, arranged in different tree structures, produce different bot behaviours? With a pre-written kernel for each function?
--- End quote ---
I was thinking more that the code in a node could change but all bots in a species would share the tree structure and code so you could go wide that way, but using preset nodes works, too.
Billy:
--- Quote from: Numsgil on April 30, 2017, 12:22:01 PM ---
--- Quote from: Billy on April 29, 2017, 08:20:50 AM ---So there would be a fixed set of these node functions that, arranged in different tree structures, produce different bot behaviours? With a pre-written kernel for each function?
--- End quote ---
I was thinking more that the code in a node could change but all bots in a species would share the tree structure and code so you could go wide that way, but using preset nodes works, too.
--- End quote ---
Why would that be easier for a GPU than DNA as it is? If the node code can mutate, wouldn't you run into the same problem of divergence?
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version