Watching the video, 11:37 [1] made me feel very uneasy about the validity of those results. CMA-ES optimization times in seconds? That only works if you do no physics, no collision checks, no trajectory roll-outs. (Otherwise you're looking at 300+ iterations with 2000+ lanes each with 1000+ physics steps each, so millions to billions of simulation steps.)
And that really makes me wonder, if 30 tokens for a robot can be so expressive that they replace 600 mio simulation steps?
Or the alternative would be that they just assume that the given trajectory is equally optimal for each robot, which to me seems like a HUUUUGE assumption. But if that's the case, then the results of this technique would be highly misleading, as they would tell you the robot is excellent for 1 example movement, while you wouldn't know that a tiny change to your example movement might make the situation much better overall with a different robot.
I think this kind of automated design process can un-hype humanoids. The point of humanoid robots is one-size-fits-all -- a generalist can readily fit into various scenarios. However, if custom designs are cheap and optimized, there's no reason to stick to inefficient generalist designs.
Let's be honest: people want humanoids because half the world was used to "domestic servants" (ie. slaves) (and even is, middle east, Pakistan, India, large parts of Africa) and they can't have that because colonialism "took that from them".
What people keep forgetting is that colonialism, and with it the large-scale disappearance of slavery, and so the disappearance of domestic servants/slaves, lasted easily into the 1990s. You don't have to be very old to remember having that (or you cheat, ie. rich middle eastern countries, and still have that in 2026, just called slightly different)
Everything from baby monitors to roombas (and the lawn equivalents) to frozen pizzas to self-driving cars to mobility scooters (replacing palanquins) to dishwashers and washing machines and tumble driers, have been chipping away at the list of tasks for which domestic servants used to be hired.
People want humanoids for the same reason they want flying cars: sci-fi made them look cooler than they would be in real life.
"The Second Renaissance" in The Animatrix justifies it rather better.
However, honestly, what happened to humans in The Matrix universe is a kindness: humanity may not know the real world, but the people still get to live out their lives as if the apocalypse (of our own doing) had never happened. The machines could've given humanity hell, not:
the first Matrix was designed to be a perfect human world, where none suffered? Where everyone would be happy? It was a disaster. No one would accept the program. Entire crops were lost. Some believed we lacked the programming language to describe your perfect world. But I believe that, as a species, human beings define their reality through misery and suffering. The perfect world was a dream that your primitive cerebrum kept trying to wake up from. Which is why the Matrix was redesigned to this, the peak of your civilization. I say your civilization because as soon as we started thinking for you, it really became our civilization
Amazing job guys. You've just created a base for robotic auto-evolution.
Now we need a model with a sole purpose of improving itself (procreation) and it's mobile platform. For the later this framework can be used. We need another model which will then do full CAD spec for each iteration and small automated factory which will implement this all in hardware ...
Watching the video, 11:37 [1] made me feel very uneasy about the validity of those results. CMA-ES optimization times in seconds? That only works if you do no physics, no collision checks, no trajectory roll-outs. (Otherwise you're looking at 300+ iterations with 2000+ lanes each with 1000+ physics steps each, so millions to billions of simulation steps.)
And that really makes me wonder, if 30 tokens for a robot can be so expressive that they replace 600 mio simulation steps?
Or the alternative would be that they just assume that the given trajectory is equally optimal for each robot, which to me seems like a HUUUUGE assumption. But if that's the case, then the results of this technique would be highly misleading, as they would tell you the robot is excellent for 1 example movement, while you wouldn't know that a tiny change to your example movement might make the situation much better overall with a different robot.
[1] https://youtu.be/TTyjvPVFbNw?si=gU8zTfUyXVIMyvyp&t=697
I'm also interested in how they figured out how to get the CMA-ES baseline to be this fast (less than 35 samples in Figure 5).
Maybe they are warm-starting CMA-ES with a pre-trained controller (although at a glance it doesn't seem they do). Haven't read the paper yet though.
[dead]
I think this kind of automated design process can un-hype humanoids. The point of humanoid robots is one-size-fits-all -- a generalist can readily fit into various scenarios. However, if custom designs are cheap and optimized, there's no reason to stick to inefficient generalist designs.
Let's be honest: people want humanoids because half the world was used to "domestic servants" (ie. slaves) (and even is, middle east, Pakistan, India, large parts of Africa) and they can't have that because colonialism "took that from them".
What people keep forgetting is that colonialism, and with it the large-scale disappearance of slavery, and so the disappearance of domestic servants/slaves, lasted easily into the 1990s. You don't have to be very old to remember having that (or you cheat, ie. rich middle eastern countries, and still have that in 2026, just called slightly different)
"Domestic servant" is much broader than slavery: https://en.wikipedia.org/wiki/Domestic_worker#Varieties_of_d...
Everything from baby monitors to roombas (and the lawn equivalents) to frozen pizzas to self-driving cars to mobility scooters (replacing palanquins) to dishwashers and washing machines and tumble driers, have been chipping away at the list of tasks for which domestic servants used to be hired.
People want humanoids for the same reason they want flying cars: sci-fi made them look cooler than they would be in real life.
("I want a flying car!" / "We have a flying car at home" / The flying car at home: https://en.wikipedia.org/wiki/Helicopter; vs "I want Brent Spiner's Data" -> https://www.youtube.com/watch?v=GvbiU_usQfE)
Looking at that last video, I have to say: what happened to humans in the matrix universe is entirely justified.
"The Second Renaissance" in The Animatrix justifies it rather better.
However, honestly, what happened to humans in The Matrix universe is a kindness: humanity may not know the real world, but the people still get to live out their lives as if the apocalypse (of our own doing) had never happened. The machines could've given humanity hell, not:
Amazing job guys. You've just created a base for robotic auto-evolution.
Now we need a model with a sole purpose of improving itself (procreation) and it's mobile platform. For the later this framework can be used. We need another model which will then do full CAD spec for each iteration and small automated factory which will implement this all in hardware ...