mlsanityData to 14 Oct 2026

DOPPLER: Dual-Policy Learning for Device Assignment in Asynchronous Dataflow Graphs

arXiv 2505.23131 ↗First posted 29 May 2025cs.LG cs.DCcited by 0

Not on the map. Why?

Abstract (arXiv)

We study the problem of assigning operations in a dataflow graph to devices to minimize execution time in a work-conserving system, with emphasis on complex machine learning workloads. Prior learning-based methods often struggle due to three key limitations: (1) reliance on bulk-synchronous systems like TensorFlow, which under-utilize devices due to barrier synchronization; (2) lack of awareness of the scheduling mechanism of underlying systems when designing learning-based methods; and (3) exclusive dependence on reinforcement learning, ignoring the structure of effective heuristics designed by experts. In this paper, we propose Doppler, a three-stage framework for training dual-policy networks consisting of 1) a mathsfSEL policy for selecting operations and 2) a mathsfPLC policy for placing chosen operations on devices. Our experiments show that Doppler outperforms all baseline methods across tasks by reducing system execution time and additionally demonstrates sampling efficiency by reducing per-episode training time.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 64 references that are in the map. First posted on arXiv 29 May 2025.

Builds on

29 of 64 references.

References · 29

  1. 2017Attention Is All You Needcited by 60,361
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
  4. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  5. 2018Attention, Learn to Solve Routing Problems!cited by 656
  6. 2021Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LMcited by 632
  7. 2016Neural Combinatorial Optimization with Reinforcement Learningcited by 608
  8. 2017Learning Combinatorial Optimization Algorithms over Graphscited by 561
  9. 2018Reinforcement Learning for Solving the Vehicle Routing Problemcited by 333
  10. 2020Reinforcement Learning for Combinatorial Optimization: A Surveycited by 236
  11. 2018Learning to Perform Local Rewriting for Combinatorial Optimizationcited by 185
  12. 2018PipeDream: Fast and Efficient Pipeline Parallel DNN Trainingcited by 117
17 more
  1. 2017Device Placement Optimization with Reinforcement Learningcited by 86
  2. 2018NVIDIA Tensor Core Programmability, Performance & Precisioncited by 79
  3. 2020Learning to Solve Combinatorial Optimization Problems on Real-World Graphs in Linear Timecited by 39
  4. 2018Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimizationcited by 27
  5. 2018Learning Permutations with Sinkhorn Policy Gradientcited by 26
  6. 2019Placeto: Learning Generalizable Device Placement Algorithms for Distributed Machine Learningcited by 26
  7. 2019Learning Heuristics over Large Graphs via Deep Reinforcement Learningcited by 23
  8. 2019Solving NP-Hard Problems on Graphs with Extended AlphaGo Zerocited by 23
  9. 2019Reinforced Genetic Algorithm Learning for Optimizing Computation Graphscited by 17
  10. 2018Improving Optimization Bounds using Machine Learning: Decision Diagrams meet Deep Reinforcement Learningcited by 13
  11. 2019Reinforcement Learning Driven Heuristic Optimizationcited by 12
  12. 2019GDP: Generalized Device Placement for Dataflow Graphscited by 10
  13. 2020Breaking (Global) Barriers in Parallel Stochastic Optimization with Wait-Avoiding Group Averagingcited by 9
  14. 2024Combinatorial Multivariant Multi-Armed Bandits with Applications to Episodic Reinforcement Learning and Beyondcited by 7
  15. 2024A Structure-Aware Framework for Learning Device Placements on Computation Graphscited by 3
  16. 2024EinDecomp: Decomposition of Declaratively-Specified Machine Learning and Numerical Computations for Parallel Executioncited by 2
  17. 2024TURNIP: A "Nondeterministic" GPU Runtime with CPU RAM Offloadcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Deft Scheduling of Dynamic Cloud Workflows with Varying Deadlines via Mixture-of-Experts
  2. 2026Cloud Workflow Scheduling Based on Graph Attention-Driven Hierarchical Reinforcement Learning
  3. 2026On the Role of DAG topology in Energy-Aware Cloud Scheduling : A GNN-Based Deep Reinforcement Learning Approach
  4. 2025GATES: Cost-aware Dynamic Workflow Scheduling via Graph Attention Networks and Evolution Strategy
  5. 2024Cost-Aware Dynamic Cloud Workflow Scheduling using Self-Attention and Evolutionary Reinforcement Learning
  6. 2026Reactive Real-Time Flow Policies via Asynchronous Distribution Alignment
  7. 2018LIFT: Reinforcement Learning in Computer Systems by Learning From Demonstrations
  8. 2025Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning