mlsanityData to 14 Oct 2026

NetRL: Powering Network Decisions via NetGRPO-Driven Reinforcement Learning

arXiv 2610.16226 ↗First posted 12 Oct 2026cs.NI cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL for energy, networks and operations

Abstract (arXiv)

Network decision-making tasks, including adaptive bitrate streaming, load balancing, and congestion control, require policies to process heterogeneous states and actions while balancing near- and long-term consequences. Existing solutions generalize poorly across network tasks, require substantial task-specific redesign, and lack reliable evaluation of long- horizon action consequences. Methods that learn through online interaction additionally face unstable optimization. We present NetRL, a two-stage LLM-based reinforcement learning framework that combines supervised fine-tuning on demonstration trajectories with online policy optimization. A universal mixed-embedding interface combines textual task descriptions with precise numerical states, while a network action head produces valid decisions. At its core, NetGRPO evaluates candidate actions over multiple future trajectories, directs deeper search toward uncertain alternatives through confidence-aware budgeting, and performs group-relative, clipped, and KL-regularized updates. NetRL achieves 23.6% higher QoE than NetLLM, 12.2% lower job completion time than Genet, and 24.1% higher network utility than Orca, yielding state-of-the-art performance among the evaluated methods. Confidence-aware budgeting reduces end-to-end training time by 27.1% without reducing validation QoE. NetRL also converges stably and generalizes to unseen network traces without retraining. Code and model weights are available at the repository linked in this paper.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for energy, networks and operations, one of 15 directions in Deep reinforcement learning methods. Builds on 13 of its 28 references that are in the map. First posted on arXiv 12 Oct 2026.

Builds on

13 of 28 references.

References · 13

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2017Deep Reinforcement Learning that Matterscited by 754
  5. 2023Secrets of RLHF in Large Language Models Part I: PPOcited by 229
  6. 2018Learning Scheduling Algorithms for Data Processing Clusterscited by 134
  7. 2024NetLLM: Adapting Large Language Models for Networkingcited by 78
  8. 2019Deep Reinforcement Learning for Autonomous Internet of Things: Model, Applications and Challengescited by 37
  9. 2016BOLA: Near-Optimal Bitrate Adaptation for Online Videoscited by 27
  10. 2025AgentRAN: An Agentic AI Architecture for Autonomous Control of Open 6G Networkscited by 16
  11. 2022Automatic Curriculum Generation for Learning Adaptation in Networkingcited by 15
  12. 2024Designing Network Algorithms via Large Language Modelscited by 12
1 more
  1. 2026LMTE: Putting the "Reasoning" into WAN Traffic Engineering with Language Modelscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026From Prior-Guided Heuristics to Deployable Agents: Accelerating Demonstration-Driven Reinforcement Learning for Deadline-Constrained Network Control
  2. 2025GEPO: Group Expectation Policy Optimization for Stable Heterogeneous Reinforcement Learning
  3. 2024Learning Sub-Second Routing Optimization in Computer Networks requires Packet-Level Dynamics
  4. 2025Dynamic Preference Multi-Objective Reinforcement Learning for Internet Network Management
  5. 2026Differentiable Policy Transport over Multi-Layer Network Feasibility Geometry
  6. 2026ThunderSyncRL: Lossless Acceleration of Agentic Reinforcement Learning
  7. 2025ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents
  8. 2026AuroraRL: Fast, Fault-Tolerant, and Cost-Efficient Reinforcement Learning over Decentralized Network