1
$\begingroup$

The problem of robotic arm control has different levels of complexity ranging from simulation to real-world application. For example, a simulation may not model friction of the joints, which becomes important in reality.

Is it possible to use a model trained in a simple environment to bootstrap learning in a more complex environment? Does this have a name? Does it fall under hierarchical reinforcement learning?

$\endgroup$
1
$\begingroup$

This concept is first presented in "Curriculum Learning" by Bengio et at.

From the abstract:

Humans and animals learn much better when the examples are not randomly presented but organized in a meaningful order which illustrates gradually more concepts, and gradually more complex ones. Here, we formalize such training strategies in the context of machine learning, and call them "curriculum learning".

$\endgroup$
  • $\begingroup$ I guess that this is not the first paper that suggest such training procedure, but it might be one of those that studies it more explicitly. It would be interesting to see papers that apply CL to the specific case of RL. Have you found anything related to RL combined with CL? $\endgroup$ – nbro May 10 at 22:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.