hierarchical-learning

Hierarchical-Learning papernotes

---

• Learning from sparse feedback from complex environments is challenging. Requires very efficient exploration strategies.

• Long-term credit assignment remains a major challenge.

Papers:

2017-10

• Meta Learning Shared Hierarchies[arXiv]

• TreeQN and ATreeC: Differentiable Tree Planning for Deep Reinforcement Learning[arXiv]

2017-06

• Learning Goal-Directed Behaviour[Pdf]

2017-05

• Feature Control as Intrinsic Motivation for Hierarchical Reinforcement Learning[arXiv]

2017-03

• Multi-Level Discovery of Deep Options[arXiv]

• FeUdal Networks for Hierarchical Reinforcement Learning[arXiv]

2016-09

• The Option-Critic Architecture[arXiv]

2016-04

• Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation[arXiv]

• A Deep Hierarchical Approach to Lifelong Learning in Minecraft [arXiv]