Welcome to my channel where I share my personal projects. You will find various experiments about many subjects like Physics Simulations, Raytracing and Machine Learning among others. Almost everything is made from scratch since I think it's the best way to learn.
I'm curious how some obstacles/terrain would affect this maybe a cluster of dots to represent a forest or a valley that prey can hide in but that is difficult to get out of
I'm interested about what would have happened if the population wasn't capped would it have ended sooner or would it have generated a more sustainable environment for both.
What's most fascinating about this is how it actually resembles different kinds of natural phenomena at the microscopic level (maybe even macro and quantum?). This is what cells have evolved to do. They change certain parameters and variables in their genes so that the next generation can be better developed. An example of one of these changes would be when you see that predators start reproducing shooting the predecessors to one side while making them move towards another direction, essentially generating a "living" barrier of predators. Of course this is very simplified and at the end we can see the AI doing some very sophisticated maneuvers, we could say that at the very end the predators are farming the prey. It really gets tricky when you then start thinking of how this all comes together in the real world for us to witness and actually for us to exist in this world. The insane amount of combined spontaneous circumstances that must have happened, and keep on happening as we try to think about it, for us and everything around us to be as it is... Having stumbled upon this knowledge makes me quite certain to say that there definitely isn't free will. There's the illusion of free will, but I guess that could be just referred to as our "minds", if that makes sense; what I mean is that we attribute to "ourselves" some traits that are in reality just an arbitrary sequence of events generated along with the electrical impulses and all the other complicated stuff going on in our bodies, and outside of them. This stuff can also be compared to how a computer works. What the computer itself does to generate this simulation, for example. Like I said earlier, these cells in the simulation are changing variables like the direction they will move next, what the generation after that one will do, and so on... Well, the computer (actually the digital memory chip, or microchip) will use voltage charges to charge different compartments within the chip to differentiate each one as a 1 or a 0. This will in turn be converted through the components of the computer, read in binary code, to then show you different kinds of pixels on your screen and very basically, everything that goes on in that screen and the reason why every single pixel is displayed as it is and every cell moves the way it does. That's how you get things like this simulation we saw (phew almost went off-topic there haha). I like techno music. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-iVfAAlCv8ec.html
What if there were a feedback loop informing individuals how their overall population was growing/shrinking, further reinforcing recent behavior? The notion that they could all know their group's status would explained by supposing, as a given, they possessed faculties of (tele?)communication (emitting a signal, like a sound)
I think that this video was focused on individual performance. Right now I see this video on my recommended next to this one called "simulating the evolution of teamwork". It may be exactly what you were looking for(?). :P ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-TZfh8hpJIxo.html
awesome video man, the graphics are super beautiful as always. you inspired me to make a network of my own, what sources did you use to learn the intricacies of the architecture? i understand the general flow but wouldn't know when at what rate should i add connections or nodes. thanks 😄
I have a question, it's not a technical one but it's regarding how you show us the output of the model, the entire rails with inverted pendulum, how did you make it?
I played the game called "Cell Lab" on Android. If you play it, you will definitely understand what is going on in this simulation, what the AI is doing. It may be a bit boring at the beginning because, at least for me, it was a bit frustrating to not understand much of it, but once you get through the basics it's truly "vaccinating" ;) , haha.
Why isn't this an ongoing series? Need a whole channel dedicated to Red v Green. Would love to see how they react and learn when there are places for ambush, hiding, resting and recovering. How about a Blue species enters the arena and we now have Rock, Paper, Scissors environment. So many awesome possibilities for learning and observation.
I wanted to do the same (create my own evolution neural network) for a long time and you gave me the motivation to ! Thank a lot Your video is really great ! Love it ! I have one question, on my side I have one issue. Network are get over complexed really fast, creating many neurons in fact not necessary. How did you managed this ? Do you decrease the chance to create a new neuron depending on the current number ? Do you take the size of the network in account when scoring ? Or maybe you didn't have this issue ! If anyone have an idea about it, I take it ! (I resolved the issue by decrasing the score depending on the network size, but I'm not feeling this is a good way to solve it)
This video was very cool I tried making this myself and my ai is just slamming the cart into the side wall and getting lucky every time because the pendulum barely goes over the top ;(
8:18 and I'm thinking that the reason the prey form big groups which then get eaten by the predator is that they spawn too quickly. Changing their viewing parameters might change things too!