Koert, Dorothea ; Maeda, Guilherme ; Neumann, Gerhard ; Peters, Jan (2022):
Learning Coupled Forward-Inverse Models with Combined Prediction Errors. (Postprint)
In: 2018 IEEE International Conference on Robotics and Automation (ICRA), pp. 2433-2439,
Darmstadt, IEEE, International Conference on Robotics and Automation (ICRA) 2018, Brisbane, QLD, Australia, 21.-25.05.2018, e-ISSN 2577-087X, ISBN 978-1-5386-3081-5,
DOI: 10.26083/tuprints-00020546,
[Conference or Workshop Item]

Abstract

Challenging tasks in unstructured environments require robots to learn complex models. Given a large amount of information, learning multiple simple models can offer an efficient alternative to a monolithic complex network. Training multiple models-that is, learning their parameters and their responsibilities-has been shown to be prohibitively hard as optimization is prone to local minima. To efficiently learn multiple models for different contexts, we thus develop a new algorithm based on expectation maximization (EM). In contrast to comparable concepts, this algorithm trains multiple modules of paired forward-inverse models by using the prediction errors of both forward and inverse models simultaneously. In particular, we show that our method yields a substantial improvement over only considering the errors of the forward models on tasks where the inverse space contains multiple solutions.

Send an inquiry

Show editorial Details