Visionary: Vision architecture discovery for robot learning

03/26/2021
by   Iretiayo Akinola, et al.
0

We propose a vision-based architecture search algorithm for robot manipulation learning, which discovers interactions between low dimension action inputs and high dimensional visual inputs. Our approach automatically designs architectures while training on the task - discovering novel ways of combining and attending image feature representations with actions as well as features from previous layers. The obtained new architectures demonstrate better task success rates, in some cases with a large margin, compared to a recent high performing baseline. Our real robot experiments also confirm that it improves grasping performance by 6 demonstrate a successful neural architecture search and attention connectivity search for a real-robot task.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset