Adaptible World Models: Learning Behaviors by Latent Imagination under Non-Stationarity

Emiliyan Gospodinov1, Vaisakh Shaj1, Philipp Becker1, Stefan Geyer2, Gerhard Neumann1,
1Autonomous Learning Robots (ALR), Karlsruhe Institute of Technology (KIT) 2Institute for Artificial Intelligence, University of Stuttgart

NeurIPS Workshop Adaptive Foundation Models

OpenReview arXiv Video Presentation Poster

Abstract

Developing foundational world models is a key research direction for embod ied intelligence, with the ability to adapt to non-stationary environments being a crucial criterion. In this work, we introduce a new formalism, Hidden Parameter-POMDP, designed for control with adaptive world models. We demonstrate that this approach enables learning robust behaviors across a variety of non-stationary RL benchmarks. Additionally, this formalism effectively learns task abstractions in an unsupervised manner, resulting in structured, task-aware latent spaces.

BibTeX

@inproceedings{gospodinov2024adaptive,
  title={Adaptive World Models: Learning Behaviors by Latent Imagination Under Non-Stationarity},
  author={Emiliyan Gospodinov and Vaisakh Shaj and Philipp Becker and Stefan Geyer and Gerhard Neumann},
  booktitle={Adaptive Foundation Models: Evolving AI for Personalized and Efficient Learning},
  year={2024},
  url={https://openreview.net/forum?id=VMIbgzRw78}
}