Improving Deep Reinforcement Learning with Knowledge Transfer

Abstract

Recent successes in applying Deep Learning techniques on Reinforcement Learning algorithms have led to a wave of breakthrough developments in agent theory and established the field of Deep Reinforcement Learning (DRL). While DRL has shown great results for single task learning, the multi-task case is still underrepresented in the available literature. This D.Sc. research proposal aims at extending DRL to the multi- task case by leveraging the power of Transfer Learning algorithms to improve the training time and results for multi-task learning. Our focus lies on defining a novel framework for scalable DRL agents that detects similarities between tasks and balances various TL techniques, like parameter initialization, policy or skill transfer.

Cite

Text

Glatt and Costa. "Improving Deep Reinforcement Learning with Knowledge Transfer." AAAI Conference on Artificial Intelligence, 2017. doi:10.1609/AAAI.V31I1.10529

Markdown

[Glatt and Costa. "Improving Deep Reinforcement Learning with Knowledge Transfer." AAAI Conference on Artificial Intelligence, 2017.](https://mlanthology.org/aaai/2017/glatt2017aaai-improving/) doi:10.1609/AAAI.V31I1.10529

BibTeX

@inproceedings{glatt2017aaai-improving,
  title     = {{Improving Deep Reinforcement Learning with Knowledge Transfer}},
  author    = {Glatt, Ruben and Costa, Anna Helena Reali},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2017},
  pages     = {5036-5037},
  doi       = {10.1609/AAAI.V31I1.10529},
  url       = {https://mlanthology.org/aaai/2017/glatt2017aaai-improving/}
}