Working Dog Forums banner

Reward training and its influences

325 Views 2 Replies 3 Participants Last post by  Mike Schoonbrood
Believe it or not, you all would have enjoyed some of the discussions I had over the last few days at a big conference with some engineers at a school out East. I interacted to a few different robotics experts, people that are designing some very awesome things, like "Leonardo." Leonardo would react when you gave it praise based on it successfully following your instructions (like opening a box, etc.).

Clicker training--or reward training--is very influential to how they design autonomous robots and improve machine learning. They were ecstatic when I showed them the "clicker training to close the kitchen drawer" video.

FWIW, autonomous robots and how they'll interact with us will be very similar to popular dog training practices today. teach, reward, teach, reward, etc. Trust me, you would have loved it.
1 - 3 of 3 Posts
:eek: That's just to wierd! Cool, but wierd.
HAL from Space Oddesy 2010 doesn't have a thing on Leonardo.
Leonardo looks like a Mogwai stuck in the Gremlin-Transitional phase. I always did want a Mogwai.
1 - 3 of 3 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.