Learning to Push by Grasping: Using multiple tasks for effective learning

09/28/2016
by   Lerrel Pinto, et al.
0

Recently, end-to-end learning frameworks are gaining prevalence in the field of robot control. These frameworks input states/images and directly predict the torques or the action parameters. However, these approaches are often critiqued due to their huge data requirements for learning a task. The argument of the difficulty in scalability to multiple tasks is well founded, since training these tasks often require hundreds or thousands of examples. But do end-to-end approaches need to learn a unique model for every task? Intuitively, it seems that sharing across tasks should help since all tasks require some common understanding of the environment. In this paper, we attempt to take the next step in data-driven end-to-end learning frameworks: move from the realm of task-specific models to joint learning of multiple robot tasks. In an astonishing result we show that models with multi-task learning tend to perform better than task-specific models trained with same amounts of data. For example, a deep-network learned with 2.5K grasp and 2.5K push examples performs better on grasping than a network trained on 5K grasp examples.

READ FULL TEXT

page 1

page 3

page 4

page 8

research
11/27/2021

GATER: Learning Grasp-Action-Target Embeddings and Relations for Task-Specific Grasping

Intelligent service robots require the ability to perform a variety of t...
research
04/12/2016

Cross-stitch Networks for Multi-task Learning

Multi-task learning in Convolutional Networks has displayed remarkable s...
research
11/05/2020

Improving Robotic Grasping on Monocular Images Via Multi-Task Learning and Positional Loss

In this paper, we introduce two methods of improving real-time object gr...
research
07/06/2017

End-to-End Learning of Semantic Grasping

We consider the task of semantic robotic grasping, in which a robot pick...
research
04/21/2022

Physics vs. Learned Priors: Rethinking Camera and Algorithm Design for Task-Specific Imaging

Cameras were originally designed using physics-based heuristics to captu...
research
06/02/2020

Learning to Branch for Multi-Task Learning

Training multiple tasks jointly in one deep network yields reduced laten...

Please sign up or login with your details

Forgot password? Click here to reset