News for hackers. Stuff that matters.
Re:Robots that learn by tomdre on Tuesday May 16, 2017 @01:23PM (#14351398)
What are the benefits of using VR vs simulation or precollected training data? Maybe I'm missing something...
Re:Robots that learn by npew on Tuesday May 16, 2017 @01:38PM (#14351553)
We actually use both. The data used to train the neural networks is all generated through a scripted policy simulation. We then use a single human demonstration in VR to show the robot how to carry out a particular task. VR lets a human do a demonstration quite naturally, and would also be an effective way to collect training data for more complex tasks, where we’re not be able to create a scripted policy.
Re:Robots that learn by tomdre on Tuesday May 16, 2017 @02:18PM (#14351957)
Thanks for your reply. So essentially VR doesn't affect the training phase per se, but simplifies the life of the trainer.
Re:Robots that learn by zardo on Tuesday May 16, 2017 @01:47PM (#14351637)
It's demonstrating one-shot learning. If you want the robot to do something that you can do, just doing it while the robot watches is a lot easier than building simulations.

It's not easy to get a million training examples, if you can learn something useful from just one example, that's great.

Re:Robots that learn by tomdre on Tuesday May 16, 2017 @02:21PM (#14351982)
Not sure this makes sense because if you can learn from one example, then you can also learn from one single simulation (assuming you have a model).
Re:Robots that learn by zardo on Tuesday May 16, 2017 @03:01PM (#14352308)
Even if you do have a model. Imagine you have the following tasks: 1) Type this sentence. 2) Specify the movements required to type that sentence in a robotics simulator.

If it can learn from 1, _you_ save a lot of effort, and the required expertise is lower.

Re:Robots that learn by JabavuAdams on Tuesday May 16, 2017 @03:19PM (#14352427)
You can generate lots and lots of simulated data. DL needs huge amounts of training data, and CG is basically a solved problem.

Actually, for me the most interesting thing is that training on synthetic images transfers so well to real camera frames. That means that I might actually be able to do some cutting edge stuff without a huge curated dataset.

Re:Robots that learn by komaromy on Tuesday May 16, 2017 @01:34PM (#14351515)
vs simulation, it lets you get trajectories from the way people actually move without having to make insanely complicated models. vs. precollected training data, maybe the recording apparatus is less expensive/complex?