Resource Management with Deep Reinforcement Learning
Explore this paper's citation graph
Summary
This work presents DeepRM, an example solution that translates the problem of packing tasks with multiple resource demands into a learning problem, and shows that it performs comparably to state-of-the-art heuristics, adapts to different conditions, converges quickly, and learns strategies that are sensible in hindsight.
- Type
- article
- Published
- 2016-11-09
- Cited by
- 1,371
- References
- 44
- OpenAlex
- https://openalex.org/W2546571074
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:207244918
Keywords
Reinforcement learning, Heuristics, Computer science, Hindsight bias, Artificial intelligence
References
- ElasticTree: Saving Energy in Data Center Networks
- Project Adam: Building an Efficient and Scalable Deep Learning Training System
- Introduction to Reinforcement Learning
- Reoptimizing Data Parallel Computing
- A Reinforcement Learning Approach to job-shop Scheduling
- Optimizing Production Manufacturing Using Reinforcement Learning
- Playing Atari with Deep Reinforcement Learning
- Neural network design
- Dominant Resource Fairness: Fair Allocation of Multiple Resource Types
- A Control-Theoretic Approach for Dynamic Adaptive Video Streaming over HTTP
- Reinforcement learning in robotics: A survey
- The tail at scale
- TCP ex machina: computer-generated congestion control
- Basis Function Adaptation in Temporal Difference Reinforcement Learning
- Quincy: fair scheduling for distributed computing clusters
- Jockey: guaranteed job latency in data parallel clusters
- Delay scheduling: a simple technique for achieving locality and fairness in cluster scheduling
- Apache Hadoop YARN: yet another resource negotiator
- Reinforcement Learning: A Survey
- Reining in the Outliers in Map-Reduce Clusters using Mantri
Cited by
- Knowledge-Defined Networking
- Deep Reinforcement Learning: An Overview
- Semisupervised Deep Reinforcement Learning in Support of IoT and Smart City Services
- o'zapft is: Tap Your Network Algorithm's Big Data!
- Lube: Mitigating Bottlenecks in Wide Area Data Analytics
- ENVI: Elastic resource flexing for Network function Virtualization
- AESOP: Automatic Policy Learning for Predicting and Mitigating Network Service Impairments
- A Machine Learning Approach to Routing
- Neural Adaptive Video Streaming with Pensieve
- A deep learning approach for optimizing content delivering in cache-enabled HetNet
- Automated Cloud Provisioning on AWS using Deep Reinforcement Learning
- Machine Learning for Networking: Workflow, Advances and Opportunities
- An Initial Investigation of Protocol Customization
- ThrottleBot - Performance without Insight
- QoS-Aware Scheduling of Heterogeneous Servers for Inference in Deep Neural Networks
- Reinforcement Learning for Primary care e Appointment Scheduling
- Learning to Route
- Harvesting Randomness to Optimize Distributed Systems
- Deep Reinforcement Learning for Multi-Resource Multi-Machine Job Scheduling
- Biases in Data-Driven Networking, and What to Do About Them
Related papers
- Addressing Hindsight Bias in Multigoal Reinforcement Learning
- Supplemental Material for Is There Hindsight Bias Without Real Hindsight? Conjectures Are Sufficient to Elicit Hindsight Bias
- Motivational influences on the strength of the hindsight bias
- Domain Knowledge and Hindsight Bias among Poker Players
- Hindsight Bias and Law
- Hindsight bias and developing theories of mind.
- HINDSIGHT BIAS IN LEGAL DECISION MAKING