Corpus ID: 219969356

Task-Agnostic Online Reinforcement Learning with an Infinite Mixture of Gaussian Processes

@article{Xu2020TaskAgnosticOR,
  title={Task-Agnostic Online Reinforcement Learning with an Infinite Mixture of Gaussian Processes},
  author={Mengdi Xu and Wenhao Ding and Jiacheng Zhu and Zuxin Liu and Baiming Chen and Ding Zhao},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.11441}
}
Continuously learning to solve unseen tasks with limited experience has been extensively pursued in meta-learning and continual learning, but with restricted assumptions such as accessible task distributions, independently and identically distributed tasks, and clear task delineations. However, real-world physical tasks frequently violate these assumptions, resulting in performance degradation. This paper proposes a continual online model-based reinforcement learning approach that does not… Expand

Figures from this paper

Accelerated Policy Evaluation: Learning Adversarial Environments with Adaptive Importance Sampling
  • Mengdi Xu, Peide Huang, +6 authors Ding Zhao
  • Computer Science
  • 2021
Functional Optimal Transport: Mapping Estimation and Domain Adaptation for Functional data

References

SHOWING 1-10 OF 52 REFERENCES
Task-Free Continual Learning
Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RL
Reconciling meta-learning and continual learning with online mixtures of tasks
Functional Regularisation for Continual Learning using Gaussian Processes
A Neural Dirichlet Process Mixture Model for Task-Free Continual Learning
Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
Multi-task reinforcement learning: a hierarchical Bayesian approach
Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation
Meta Reinforcement Learning with Latent Variable Gaussian Processes
...
1
2
3
4
5
...