[10] Chelsea Finn - Learning to Learn with Gradients
The Thesis Review
00:00
The Future of Artificial Intelligence and Robotics
I think that if we can ultimately put agents in the real world, then that may make these algorithms more practical. And as the complexity of that world increases, also I think the complexity of the behavior that's learned will also have the potential to increase. It actually kind of bugs me a lot when we run a lot of experiments on really simple worlds because the only thing that you'll be able to get out of very, very simple worlds is very simple behaviors.
Transcript
Play full episode
