Learning Coupled Forward-Inverse Models with Combined Prediction Errors
Lists
Only you see this
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.
BibTeX
@inproceedings{koert2018learning,
author = {Dorothea Koert and Guilherme Maeda and Gerhard Neumann and Jan Peters},
title = {{Learning Coupled Forward-Inverse Models with Combined Prediction Errors}},
booktitle = {IEEE International Conference on Robotics and Automation},
year = {2018},
url = {https://openreview.net/forum?id=yvFh6dv8N-j},
}Loading related content...