Skip to main content Skip to main navigation


Learning Coupled Forward-Inverse Models with Combined Prediction Errors

Dorothea Koert; Guilherme Maeda; Gerhard Neumann; Jan Peters
In: 2018 IEEE International Conference on Robotics and Automation. IEEE International Conference on Robotics and Automation (ICRA-2018), May 21-25, Brisbane, Australia, Pages 2433-2439, IEEE, 2018.


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.

Weitere Links