Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models

no code implementations • 3 Aug 2023 • Joao Carvalho, An T. Le, Mark Baierl, Dorothea Koert, Jan Peters

Learning priors on trajectory distributions can help accelerate robot motion planning optimization.

Denoising Motion Planning

Diminishing Return of Value Expansion Methods in Model-Based Reinforcement Learning

1 code implementation • 7 Mar 2023 • Daniel Palenicek, Michael Lutter, Joao Carvalho, Jan Peters

Therefore, we conclude that the limitation of model-based value expansion methods is not the model accuracy of the learned models.

Continuous Control Model-based Reinforcement Learning +2

An Analysis of Measure-Valued Derivatives for Policy Gradients

no code implementations • 8 Mar 2022 • Joao Carvalho, Jan Peters

This estimator is unbiased, has low variance, and can be used with differentiable and non-differentiable function approximators.

A Nonparametric Off-Policy Policy Gradient

1 code implementation • 8 Jan 2020 • Samuele Tosatto, Joao Carvalho, Hany Abdulsamad, Jan Peters

Reinforcement learning (RL) algorithms still suffer from high sample complexity despite outstanding recent successes.

Density Estimation Policy Gradient Methods +1