Active inverse model learning with error and reachable set estimates

D Driess, S Schmitt, M Toussaint - 2019 IEEE/RSJ International …, 2019 - ieeexplore.ieee.org
In this work, we propose a framework to learn an inverse model of redundant systems. We
address three problems. By formalizing what it actually means to learn an inverse model, we …

Leveraging forward model prediction error for learning control

S Bechtle, B Hammoud, A Rai, F Meier… - … on Robotics and …, 2021 - ieeexplore.ieee.org
Learning for model based control can be sample-efficient and generalize well, however
successfully learning models and controllers that represent the problem at hand can be …

Lifelong learning in the real world

SME Bechtle - 2022 - tobias-lib.ub.uni-tuebingen.de
For robots to assist and support humans in their daily lives, they will have to learn from-and
adapt to our world. But, applying machine learning to real world problems is still very …

[PDF][PDF] Efficient and stable online learning for developmental robots

R Rayyes - 2020 - scholar.archive.org
Efficient and Stable Online Learning for Developmental Robots Page 1 Efficient and Stable
Online Learning for Developmental Robots Von der Carl-Friedrich-Gauß-Fakultät der …

[引用][C] Improving Multi-Step Model-Based Reinforcement Learning

NJ Jie