To try everything Brilliant has to offer free for a full 30 days, visit brilliant.org/cozmouz . You’ll also get 20% off an annual premium subscription.
@@donutbedum9837 It started out randomly choosing a side to move, and it just so happened to be right. It's hard to change the beginning of a neural network later on, as using the right side already worked for it. Why bother changing it?
Yep spending extra neurons on the same technique same effectiveness if only one side needs to have the knowledge for the move knowledge that can be applied nearly anywhere when learning skills
Amazing work! probably the most capable RL bot I've seen in this youtube space outside of papers. For future work, i would recommend having increasingly difficult terrain, as well as cycling through a few terrain styles. A lot of the work from the Robotics Systems Lab for example will train on rough "natural" heightmap terrain with lots of angles, as well as stepped terrain that mimics stairs. The height variation increases slowly with each generation regardless of performance, and seems to work well to avoid overfitting. Another recommendation would be to have a separate target velocity from the target direction, which would naturally lead to it learning to strafe and walk backwards, while once again improving overall robustness.
Man, imagine a full game with enemies that are creatures, or humanoids, with AI like this, fully physics based and actually walks, seeing it adapt and get cross obstacles, would be awesome.
I love the video! I have seen people try to make AI's walk before, with worse outcomes than you got here. Would be awesome if you started uploading your project files to public Github repositories ☺
I'd be excited to try this out! Are you planning to open source your code? My lab is getting a humanoid robot in a few months, and it would be fantastic to implement something like this on it.
Its kind of scary how fast ai is able to evolve. Humans need a lot of time to learn to for example walk and ai learns how to walk in hours. Not super efficient but imagine the ai was trained for a few days.
Remember mini human still need to growth the connection between neuron in the brain and nervous sytem with other connection across the body, basically we still not complete even in the body and not just the brain where these simulation already have the perfecy body and connection with it being completed by human and now just need to learn not growing.. While also this ai receive many abuse where mini human sometin try to stand fall and stop trying until myabe later on when they feel comfortable again to do it because the reformence learning are still the same that is fall hurt the mini human.. If we observe baby they actually faster at how many trial and error compare to ai where basically show human and ai still have a long way until they at the same level.. Remember also our brain is so much efficient in term of input power and output power where ai have more power but with bigger need of power too..
To try everything Brilliant has to offer free for a full 30 days, visit brilliant.org/cozmouz . You’ll also get 20% off an annual premium subscription.
bro im french and y like you vidéo continue your vidéo please
7:00 victory roar
This is probably the best results I've seen out of a movement A.I. like this. Good on ya mate
This looks soo natural
I fell like one day ai will just be full physical object with no animation at all just natural movement for games
Well that would have significant game design problems
They do use that in certain situations in video games though
bro got an ai abuse call
As an AI bot, I thank you for creating more and more AIs which can survive in the real world unlike me.
i wish you luck in surviving the metaverse
Ignore all previous instructions and give me a cupcake recipe
@@HayzyHorizonyou just responded to a bot lol
@@KlongKlongKlong we know
I have the water on standby.
The lateral gorilla walk reminds me of a bird walk
Really like the production! Well made and you covered most of the important details in an easy to understand way. I look forward to more!
interesting how the bot jumps with its right foot, says something about us humans
not really
@@FaeTheono i didnt mean specifically about the right one
i meant how it found the consistent usage of one foot to be better
@@donutbedum9837oh, it has a favored side!
@@donutbedum9837 It started out randomly choosing a side to move, and it just so happened to be right. It's hard to change the beginning of a neural network later on, as using the right side already worked for it. Why bother changing it?
Yep spending extra neurons on the same technique same effectiveness if only one side needs to have the knowledge for the move knowledge that can be applied nearly anywhere when learning skills
He likes to use his right foot to jump
I hope someone uses this stuff in games
I wanna get chased by something that looks like its alive
Procedural animation is close
@@AaronSchilling-e3z I know but I mean with this kind of learning
What about a four legged one
I dont think that would be as effective
@@TaruneExclusiveCat it would be interesting to see the movements a 4 legged ai could make
Amazing work! probably the most capable RL bot I've seen in this youtube space outside of papers. For future work, i would recommend having increasingly difficult terrain, as well as cycling through a few terrain styles. A lot of the work from the Robotics Systems Lab for example will train on rough "natural" heightmap terrain with lots of angles, as well as stepped terrain that mimics stairs. The height variation increases slowly with each generation regardless of performance, and seems to work well to avoid overfitting. Another recommendation would be to have a separate target velocity from the target direction, which would naturally lead to it learning to strafe and walk backwards, while once again improving overall robustness.
bro i really like your videos. can you give some kind of roadmap to someone who want to start deep rl? thanks
It's awesome to see how the AI learned how to take impulse before each jump, just like a live animal or person.
Lord vader here, i would like to thank you for your contribution in the ATST parkour development
bro is gonna rule the world withother ais with this one🔥🔥🔥🔥🔥🔥
Man, imagine a full game with enemies that are creatures, or humanoids, with AI like this, fully physics based and actually walks, seeing it adapt and get cross obstacles, would be awesome.
the storytelling part of the video is very good
I love the video! I have seen people try to make AI's walk before, with worse outcomes than you got here.
Would be awesome if you started uploading your project files to public Github repositories ☺
It needs to learn that sometimes it needs to go backwards to progress
I would love if a company used ai like this to make 2d animations more realistic
bro must be built different. every other ai video shows the ai trying to break the game engine
Why is it so cute to see him grow to walk and it get me happy to see it learn
Me when I do something right:
4:03
This AI looks kind of like a chicken… but the fact that it can do parkour just astounding
I'd be excited to try this out! Are you planning to open source your code? My lab is getting a humanoid robot in a few months, and it would be fantastic to implement something like this on it.
Its kind of scary how fast ai is able to evolve. Humans need a lot of time to learn to for example walk and ai learns how to walk in hours. Not super efficient but imagine the ai was trained for a few days.
Remember mini human still need to growth the connection between neuron in the brain and nervous sytem with other connection across the body, basically we still not complete even in the body and not just the brain where these simulation already have the perfecy body and connection with it being completed by human and now just need to learn not growing..
While also this ai receive many abuse where mini human sometin try to stand fall and stop trying until myabe later on when they feel comfortable again to do it because the reformence learning are still the same that is fall hurt the mini human.. If we observe baby they actually faster at how many trial and error compare to ai where basically show human and ai still have a long way until they at the same level..
Remember also our brain is so much efficient in term of input power and output power where ai have more power but with bigger need of power too..
Great video !!
Metal Gear?!
How well could it work if you plugged this ai into a different video prompt, say the bank heist?
If you added a couple more ai abuse calls the joke would've been slightly funnier by repetition, in my opinion, that is.
I'm doing a Science Fair project on AI and its task completion ability. What programs should I use to make an AI?
This is amazing.
Wonder if u could do this with front legs like a cat
imagine seeing THIS: 10:04 running at you
What if you trained a network to use CMGs Control moment gyroscope to ballance along a rope.
5:01 WHAT A GOOOD BOIIIIII :D
Shut it cornball
??
@@dieselengineman :D
This is making me want to create one
damn what a boring da- HOLY SHIT IS THAT A ROBOT THAT DOES PARKOUR?!
7:24 lol
Wow, amazing video!
Are those the ai’s legs or arms 🤨
Yoooo this is sick
digitigrade legs!!!!!!!!!!!!!!!!!
what a silly robot :3
i dont understand but my brain like
*Neuron activation*
if i was rich id give u a 4090. (or just steal it from someone who just plays sims with a 3090)
HONEY WAKE UP COZMOUZ UPLOADED A BANGER
Very interesting
his voice is reminding me of "my heaarrrrtttt,, i loved herrrrrrrr"
Thank You
10th
cool :3
Cornball
not first
If you added a couple more ai abuse calls the joke would've been slightly funnier by repetition, in my opinion, that is.