Hierarchical Policy Learning Is Sensitive To Goal Space Design

Hierarchy in reinforcement learning agents allows for control at multiple time scales yielding improved sample efficiency, the ability to deal with long time horizons and transferability of sub-policies to tasks outside the training distribution.

Authors

Mariano Phielipp

Senior Deep Learning Data Scientist, Intel AI Lab

View authors bio

Arjun Bansal

Vice President and General Manager, Artificial Intelligence Software and Lab at Intel

View authors bio

Zach Dwiel

Senior Algorithms Engineer, Intel AI Lab

Related Content

Label Efficient Audio Classification Through Multitask Learning And...

While deep learning has been incredibly successful in modeling tasks with large, carefully curated labeled datasets, its application to problems...

View publication

Heuristics For Image Generation From Scene Graphs

Generating realistic images from scene graphs requires neural networks to be able to reason about object relationships and compositionality. Learning...

View publication

Graph-DQN: Fast Generalization To Novel Objects Using Prior…

Humans have a remarkable ability to both generalize known actions to novel objects, and reason about novel objects once their..

View publication

SPIGAN: Privileged Adversarial Learning from Simulation

Deep Learning for Computer Vision depends mainly on the source of supervision. Photo-realistic simulators can generate large-scale automatically labeled synthetic...

View publication

Stay Connected


Keep tabs on all the latest news with our monthly newsletter.