RLlib: Abstractions for Distributed Reinforcement Learning
Explore this paper's citation graph
Summary
This work argues for distributing RL components in a composable way by adapting algorithms for top-down hierarchical control, thereby encapsulating parallelism and resource requirements within short-running compute tasks, through RLlib: a library that provides scalable software primitives for RL.
- Type
- article
- Published
- 2017-12-26
- Cited by
- 1,059
- References
- 42
- Access
- Open access
- OpenAlex
- https://openalex.org/W2779040504
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:49546141
Keywords
Composability, Computer science, Scalability, Reinforcement learning, Reuse
References
- Bigtable: a distributed storage system for structured data
- The tail at scale
- The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines
- A High-Performance, Portable Implementation of the MPI Message Passing Interface Standard
- Scaling Distributed Machine Learning with the Parameter Server
- Encapsulation of Parallelism and Architecture-Independence in Extensible Database Query Execution
- Human-level control through deep reinforcement learning
- Caffe: Convolutional Architecture for Fast Feature Embedding
- Composing parallel software efficiently with lithe
- MapReduce: simplified data processing on large clusters
- MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
- Spark: Cluster Computing with Working Sets
- TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
- Distributed TensorFlow with MPI
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning
- Proximal Policy Optimization Algorithms
- Hyperband: Bandit-Based Configuration Evaluation for Hyperparameter Optimization
- TensorFlow Agents: Efficient Batched Reinforcement Learning in TensorFlow
- Mastering the game of Go without human knowledge
- Automatic differentiation in PyTorch
Cited by
- Productivity, Portability, Performance: Data-Centric Python
- RLgraph: Flexible Computation Graphs for Deep Reinforcement Learning
- Dopamine: A Research Framework for Deep Reinforcement Learning
- Iroko: A Framework to Prototype Reinforcement Learning for Data Center Traffic Control
- Catalyst.RL: A Distributed Framework for Reproducible RL Research
- Reinforcement Learning vs. Backstepping Control of Stop-and-Go Traffic
- Attentional Policies for Cross-Context Multi-Agent Reinforcement Learning
- Reinforcement Learning for Channel Coding: Learned Bit-Flipping Decoding
- Towards White-box Benchmarks for Algorithm Control
- Deep Reinforcement Learning: Frontiers of Artificial Intelligence
- Learning Safe Unlabeled Multi-Robot Planning with Motion Constraints
- Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning
- Deep Reinforcement Learning in Match-3 Game
- simple_rl: Reproducible Reinforcement Learning in Python
- Neural packet classification
- A Reinforcement Learning Approach for Control of a Nature-Inspired Aerial Vehicle
- Reinforcement Learning Agent under Partial Observability for Traffic Light Control in Presence of Gridlocks
- A review of cooperative multi-agent deep reinforcement learning
- A Unified Bellman Optimality Principle Combining Reward Maximization and Empowerment
- Differentiable Cloth Simulation for Inverse Problems
Related papers
- Proximal Policy Optimization Algorithms
- RLlib: Abstractions for Distributed Reinforcement Learning
- Human-level control through deep reinforcement learning
- Trust Region Policy Optimization
- Playing Atari with Deep Reinforcement Learning
- Mastering the game of Go with deep neural networks and tree search
- Distributed Prioritized Experience Replay