# Teal: Learning-Accelerated Optimization of WAN Traffic Engineering

arXiv: https://arxiv.org/abs/2210.13763 · HTML: https://mlsanity.com/paper/2210.13763/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 77 references that are in the map.
- First posted on arXiv 25 Oct 2022.

## Builds on (most cited first, 16 of 16)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Neural Message Passing for Quantum Chemistry](https://mlsanity.com/paper/1704.01212/) (2017)
- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [Counterfactual Multi-Agent Policy Gradients](https://mlsanity.com/paper/1705.08926/) (2017)
- [Graph Neural Networks for Social Recommendation](https://mlsanity.com/paper/1902.07243/) (2019)
- [Learning to Communicate with Deep Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1605.06676/) (2016)
- [Representation Learning on Graphs: Methods and Applications](https://mlsanity.com/paper/1709.05584/) (2017)
- [Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Prediction](https://mlsanity.com/paper/2002.11927/) (2020)
- [Caffe: Convolutional Architecture for Fast Feature Embedding](https://mlsanity.com/paper/1408.5093/) (2014)
- [cuDNN: Efficient Primitives for Deep Learning](https://mlsanity.com/paper/1410.0759/) (2014)
- [Solving Mixed Integer Programs Using Neural Networks](https://mlsanity.com/paper/2012.13349/) (2020)
- [Experience-driven Networking: A Deep Reinforcement Learning based Approach](https://mlsanity.com/paper/1801.05757/) (2018)
- [Solving Large-Scale Granular Resource Allocation Problems Efficiently with POP](https://mlsanity.com/paper/2110.11927/) (2021)
- [CFR-RL: Traffic Engineering with Reinforcement Learning in SDN](https://mlsanity.com/paper/2004.11986/) (2020)
- [Dynamic Graph Neural Network for Traffic Forecasting in Wide Area Networks](https://mlsanity.com/paper/2008.12767/) (2020)

## Built on by (most cited first, 13 of 13)

- [Diffusion Models as Network Optimizers: Explorations and Analysis](https://mlsanity.com/paper/2411.00453/) (2024)
- [Near-Real-Time Resource Slicing for QoS Optimization in 5G O-RAN using Deep Reinforcement Learning](https://mlsanity.com/paper/2509.14343/) (2025)
- [Learning Sub-Second Routing Optimization in Computer Networks requires Packet-Level Dynamics](https://mlsanity.com/paper/2410.10377/) (2024)
- [iScheduler: Reinforcement Learning-Driven Continual Optimization for Large-Scale Resource Investment Problems](https://mlsanity.com/paper/2602.06064/) (2026)
- [EMA: Efficient Model Adaptation for Learning-based Systems](https://mlsanity.com/paper/2605.13942/) (2026)
- [Introduction to Graph Neural Networks for Machine Learning Engineers](https://mlsanity.com/paper/2412.19419/) (2024)
- [Traffic Engineering in Large-scale Networks with Generalizable Graph Neural Networks](https://mlsanity.com/paper/2503.24203/) (2025)
- [Geminet: Learning the Duality-based Iterative Process for Lightweight Traffic Engineering in Changing Topologies](https://mlsanity.com/paper/2506.23640/) (2025)
- [From GNNs to Symbolic Surrogates via Kolmogorov-Arnold Networks for Delay Prediction](https://mlsanity.com/paper/2512.20885/) (2025)
- [Can Neural Networks Provide Latent Embeddings for Telemetry-Aware Greedy Routing?](https://mlsanity.com/paper/2602.12798/) (2026)
- [Towards Near-Real-Time Telemetry-Aware Routing with Neural Routing Algorithms](https://mlsanity.com/paper/2604.02927/) (2026)
- [Analyzing Symbolic Properties for DRL Agents in Systems and Networking](https://mlsanity.com/paper/2604.04914/) (2026)
- [NeuroRisk: Physics-Informed Neural Optimization for Risk-Aware Traffic Engineering](https://mlsanity.com/paper/2605.12862/) (2026)

## Abstract (arXiv)

The rapid expansion of global cloud wide-area networks (WANs) has posed a challenge for commercial optimization engines to efficiently solve network traffic engineering (TE) problems at scale. Existing acceleration strategies decompose TE optimization into concurrent subproblems but realize limited parallelism due to an inherent tradeoff between run time and allocation performance. We present Teal, a learning-based TE algorithm that leverages the parallel processing power of GPUs to accelerate TE control. First, Teal designs a flow-centric graph neural network (GNN) to capture WAN connectivity and network flows, learning flow features as inputs to downstream allocation. Second, to reduce the problem scale and make learning tractable, Teal employs a multi-agent reinforcement learning (RL) algorithm to independently allocate each traffic demand while optimizing a central TE objective. Finally, Teal fine-tunes allocations with ADMM (Alternating Direction Method of Multipliers), a highly parallelizable optimization algorithm for reducing constraint violations such as overutilized links. We evaluate Teal using traffic matrices from Microsoft's WAN. On a large WAN topology with >1,700 nodes, Teal generates near-optimal flow allocations while running several orders of magnitude faster than the production optimization engine. Compared with other TE acceleration schemes, Teal satisfies 6--32% more traffic demand and yields 197--625x speedups.