Bayesian Optimization With Automatic Prior Selection For Dataefficient Direct Policy Search

# Bayesian Optimization with Automatic Prior Selection for Data-Efficient Direct Policy Search

**arXiv ID:** 1709.06919
**Authors:** Rémi Pautrat, Konstantinos Chatzilygeroudis, Jean-Baptiste Mouret
**Published:** 2017-09-20T15:04:50Z
**Abstract:**
One of the most interesting features of Bayesian optimization for direct policy search is that it can leverage priors (e.g., from simulation or from previous tasks) to accelerate learning on a robot. In this paper, we are interested in situations for which several priors exist but we do not know in advance which one fits best the current situation. We tackle this problem by introducing a novel acquisition function, called Most Likely Expected Improvement (MLEI), that combines the likelihood of the priors and the expected improvement. We evaluate this new acquisition function on a transfer learning task for a 5-DOF planar arm and on a possibly damaged, 6-legged robot that has to learn to walk on flat ground and on stairs, with priors corresponding to different stairs and different kinds of damages. Our results show that MLEI effectively identifies and exploits the priors, even when there is no obvious match between the current situations and the priors.

## Skill Description

This skill is generated from the arXiv paper: Bayesian Optimization with Automatic Prior Selection for Data-Efficient Direct Policy Search (1709.06919).

## How to Use

[To be filled in by the user or by future automation]

## References

- [arXiv:1709.06919](http://arxiv.org/abs/1709.06919v2)