Publication

Mutual information weighing for probabilistic movement primitives

Conference Article

Conference

Catalan Conference on Artificial Intelligence (CCIA)

Edition

24th

Pages

365-368

Doc link

https://doi.org/10.3233/FAIA220359

File

Download the digital copy of the doc pdf document

Abstract

Reinforcement Learning (RL) of trajectory data has been used in several fields, and it is of relevance in robot motion learning, in which sampled trajectories are run and their outcome is evaluated with a reward value. The responsibility on the performance of a task can be associated to the trajectory as a whole, or distributed throughout its points (timesteps). In this work, we present a novel method for attributing the responsibility of the rewards to each time-step separately by using Mutual Information (MI) to bias the model fitting of a trajectory.

Categories

artificial intelligence, robots, stochastic programming.

Scientific reference

A. Colomé and C. Torras. Mutual information weighing for probabilistic movement primitives, 24th Catalan Conference on Artificial Intelligence, 2022, Sitges, in Artificial Intelligence Research and Development, Vol 356 of Frontiers in Artificial Intelligence and Applications, pp. 365-368, IOS Press.