Multi-Instance Aware Localization for End-to-End Imitation Learning

12/26/2020
by   Sagar Gubbi Venkatesh, et al.
0

Existing architectures for imitation learning using image-to-action policy networks perform poorly when presented with an input image containing multiple instances of the object of interest, especially when the number of expert demonstrations available for training are limited. We show that end-to-end policy networks can be trained in a sample efficient manner by (a) appending the feature map output of the vision layers with an embedding that can indicate instance preference or take advantage of an implicit preference present in the expert demonstrations, and (b) employing an autoregressive action generator network for the control layers. The proposed architecture for localization has improved accuracy and sample efficiency and can generalize to the presence of more instances of objects than seen during training. When used for end-to-end imitation learning to perform reach, push, and pick-and-place tasks on a real robot, training is achieved with as few as 15 expert demonstrations.

READ FULL TEXT
research
05/18/2018

Reinforced Imitation: Sample Efficient Deep Reinforcement Learning for Map-less Navigation by Leveraging Prior Demonstrations

This work presents a learning-based approach for target driven map-less ...
research
05/29/2019

Adversarial Imitation Learning from Incomplete Demonstrations

Imitation learning targets deriving a mapping from states to actions, a....
research
10/06/2017

End-to-end Driving via Conditional Imitation Learning

Deep networks trained on demonstrations of human driving have learned to...
research
04/06/2023

End-to-end Manipulator Calligraphy Planning via Variational Imitation Learning

Planning from demonstrations has shown promising results with the advanc...
research
07/29/2020

Sample Efficient Interactive End-to-End Deep Learning for Self-Driving Cars with Selective Multi-Class Safe Dataset Aggregation

The objective of this paper is to develop a sample efficient end-to-end ...
research
12/26/2020

Stochastic Action Prediction for Imitation Learning

Imitation learning is a data-driven approach to acquiring skills that re...
research
09/16/2019

Self-Supervised Correspondence in Visuomotor Policy Learning

In this paper we explore using self-supervised correspondence for improv...

Please sign up or login with your details

Forgot password? Click here to reset