Wield: Systematic Reinforcement Learning With Progressive Randomization
arXiv:1909.06844
Abstract
Reinforcement learning frameworks have introduced abstractions to implement and execute algorithms at scale. They assume standardized simulator interfaces but are not concerned with identifying suitable task representations. We present Wield, a first-of-its kind system to facilitate task design for practical reinforcement learning. Through software primitives, Wield enables practitioners to decouple system-interface and deployment-specific configuration from state and action design. To guide experimentation, Wield further introduces a novel task design protocol and classification scheme centred around staged randomization to incrementally evaluate model capabilities.
arXiv comment10 pages, draft paper
References in corpus (13)
- Unity: A General Platform for Intelligent Agents
- Neo: A Learned Query Optimizer
- Challenges of Real-World Reinforcement Learning
- Deep Reinforcement Learning for Join Order Enumeration
- Device Placement Optimization with Reinforcement Learning
- Quantifying Generalization in Reinforcement Learning
- Simple random search provides a competitive approach to reinforcement learning
- Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics
- AI Safety Gridworlds
- ns3-gym: Extending OpenAI Gym for Networking Research
- A Benchmark Environment Motivated by Industrial Control Problems
- Placeto: Learning Generalizable Device Placement Algorithms for Distributed Machine Learning
- A Machine Learning Approach to Routing