this post was submitted on 23 Nov 2023
1 points (100.0% liked)

Machine Learning

1 readers
1 users here now

Community Rules:

founded 1 year ago
MODERATORS
 

I've spent the past few months diving into recent publications from conferences like CoRL, ICRA, IROS, RAUL, CVPR, and some preprints of note. It seems like we're getting pretty close to robotic assistants that can perform a limited range of tasks and be deployed in people's homes. Where do we think we are with this tech? What do you think is the biggest bottleneck we need to overcome? Do you think the "internet scale" behavioural cloning techniques from Google will lead the way? Or something more RL-VL oriented?

I'm at the start of an AI PhD with a focus on robotics so I'd love to hear people's thoughts on this!

you are viewing a single comment's thread
view the rest of the comments
[–] theLanguageSprite@alien.top 1 points 11 months ago (1 children)

Well we already have roombas and robots that serve tables at restaurants, but I'm guessing you mean more like a robot butler from Fallout, capable of receiving verbal instructions and performing arbitrary tasks. As far as I can tell, there are two main reasons.

The first is that robots are dangerous. If you make a robot strong enough to perform tasks that humans do, like lifting heavy objects, climbing stairs, and opening doors, it's gonna have to be pretty heavy. There are tons of scenarios where the robot could fall on someone, drop something on someone, or just hit them with something really fast. The best way to overcome this would probably be one of those baymax style squishy robots, but it still seems like a challenging engineering problem.

The second is more the ML side of things. Even with "a limited range of tasks", you correctly pointed out you'd need RL, because every environment is different. If you're curious about why RL hasn't seen many real world applications yet, I'd recommend you try to train an RL agent yourself. Even with the state of the art, it's incredibly mercurial, and often falls into a failure mode unless the hyperparameters are just right and the observations and rewards are designed perfectly. The people at nvidia, meta, and openai are desperately trying to monetize RL with all sorts of tricks and techniques, but it's just not there yet.

So ultimately we'll probably have robot butlers around the time we solve both of these problems, which depending on who you ask, will either be in the next couple decades or never.

[–] VAL9THOU@alien.top 1 points 11 months ago (1 children)

There's also the whole "if your robot is smart enough to run your house for you, with all of the decision making ability and contextual analysis that would entail, it's probably smart enough that owning it should probably be considered slavery" part

[–] perta1234@alien.top 1 points 11 months ago

Maybe, though we would have no problem if a dog was able to do that? Anyway, it could be a set of specialized skills without general skills There could be appearance of intelligence without actual general intelligence