Robust and Efficient Transfer Learning with Hidden-Parameter Markov Decision Processes
Machine Learning
2017-11-01 v3 Artificial Intelligence
Machine Learning
Abstract
We introduce a new formulation of the Hidden Parameter Markov Decision Process (HiP-MDP), a framework for modeling families of related tasks using low-dimensional latent embeddings. Our new framework correctly models the joint uncertainty in the latent parameters and the state space. We also replace the original Gaussian Process-based model with a Bayesian Neural Network, enabling more scalable inference. Thus, we expand the scope of the HiP-MDP to applications with higher dimensions and more complex dynamics.
Cite
@article{arxiv.1706.06544,
title = {Robust and Efficient Transfer Learning with Hidden-Parameter Markov Decision Processes},
author = {Taylor Killian and Samuel Daulton and George Konidaris and Finale Doshi-Velez},
journal= {arXiv preprint arXiv:1706.06544},
year = {2017}
}
Comments
To appear at NIPS 2017, selected for an oral presentation. 17 pages (incl references and appendix). Example code can be found at http://github.com/dtak/hip-mdp-public