RLgraph: Modular Computation Graphs for Deep Reinforcement Learning
Accepted version
Peer-reviewed
Repository URI
Repository DOI
Change log
Authors
Abstract
Reinforcement learning (RL) tasks are challenging to implement, execute and test due to algorithmic instability, hyper-parameter sensitivity, and heterogeneous distributed communication patterns. We argue for the separation of logical component composition, backend graph definition, and distributed execution. To this end, we introduce RLgraph, a library for designing and executing reinforcement learning tasks in both static graph and define-by-run paradigms. The resulting implementations are robust, incrementally testable, and yield high performance across different deep learning frameworks and distributed backends.
Description
Keywords
cs.LG, cs.LG, cs.AI, stat.ML
Journal Title
CoRR
Conference Name
SysML: Conference on Systems and Machine Learning
Journal ISSN
Volume Title
Publisher
Publisher DOI
Rights
All rights reserved