AMO: Adaptive Muon Orthogonalization

• URL: http://arxiv.org/abs/2605.17806v1
• Date: Mon, 18 May 2026 03:29:18 GMT
• Title: AMO: Adaptive Muon Orthogonalization
• Abstract summary: We show that per-matrix heterogeneity is pervasive and largely determined by matrix geometry.<n>We propose Adaptive Muon Orthogonalization (AMO), an observe-then-commit method that measures weight geometry by operator type early in training.<n>AMO delivers consistent improvements over uniform-schedule Muon across standard, prolonged, and continual pre-training.
• Score: 28.545002803615606
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Muon has recently emerged as a competitive alternative to AdamW for large-scale pre-training, with orthogonalization via Newton-Schulz (NS) iterations as its core operation. Existing Muon variants apply a uniform NS schedule to all parameter matrices, overlooking possible differences in orthogonalization difficulty and its impact on performance. Through a systematic empirical study, we show that this per-matrix heterogeneity is pervasive and largely determined by matrix geometry, which evolves dynamically across operator types, training stages, and network depths. As a result, uniform NS schedules can lead to uneven orthogonalization quality across the model. Motivated by these findings, we propose Adaptive Muon Orthogonalization (AMO), an observe-then-commit method that measures weight geometry by operator type early in training and then uses these signals to allocate the NS budget for the remainder of training. AMO delivers consistent improvements over uniform-schedule Muon across standard, prolonged, and continual pre-training, surpassing the strongest baseline by +0.76 on Llama3.1-1.4B and +0.51 on Qwen3-1.7B in average downstream performance of 12 evaluation tasks.

Related papers

• Improving Neural Network Training by Decoupling the Magnitude and Direction of Weight Vectors [110.10775872372047]
  We propose emphMagnitude-Direction-Direction (MD) Decoupling.<n>It factorizes each weight into a fixed-norm direction on a hypersphere and learnable per-row and per-column magnitude gains, updated at separate learning rates.<n>Across both Adam and Muon, MD Decoupling improves on well-tuned baselines, transfers the optimal LR across model width without retuning, and continues to help at scale on large Mixture-of-Experts models.
  arXiv  Detail & Related papers  (2026-06-24T15:40:26Z)
• Adam Improves Muon: Adaptive Moment Estimation with Orthogonalized Momentum [5.049533819651459]
  We propose a new and a diagonal extension, NAMO and NAMO-D, to provide the first principled integration of momentum with noise adaptation.<n> NAMO-D instead right-multiplies orthogonalized momentum by a diagonal matrix with clamped entries.<n>Experiments on pretraining GPT-2 models demonstrate improved performance of both NAMO and NAMO-D compared to the AdamW and Muon models.
  arXiv  Detail & Related papers  (2026-02-19T05:00:39Z)
• Variance-Adaptive Muon: Accelerating LLM Pretraining with NSR-Modulated and Variance-Scaled Momentum [19.385264518362472]
  Large Language Models (LLMs) achieve competitive performance across diverse natural language processing (NLP) tasks.<n>We propose two variants of Muon, Muon-NSR and Muon-VS, which apply variance-adaptive normalization to momentum.<n>Experiments on GPT-2 and LLaMA pretraining demonstrate that our proposed methods accelerate convergence and consistently achieve lower validation loss than both competitive, well-tuned AdamW and Muon baselines.
  arXiv  Detail & Related papers  (2026-01-21T02:41:56Z)
• MuonBP: Faster Muon via Block-Periodic Orthogonalization [24.232069944820513]
  We show how to adjust the learning rate from the baseline to MuonBP and give guarantees for this algorithm.<n>When training an 8B model with eight-way tensor tensor and ZeRO statewiseing, MuonBP achieves 8% Muon with no degradation in performance.
  arXiv  Detail & Related papers  (2025-10-19T19:56:05Z)
• NorMuon: Making Muon more efficient and scalable [71.49702449498085]
  We propose NorMuon (Neuron-wise Normalized Muon) as a successor to Adam.<n>We show NorMuon consistently outperforms both Adam and Muon, achieving 21.74% better training efficiency than Adam and 11.31% improvement over Muon on 1.1 B pretraining setting.
  arXiv  Detail & Related papers  (2025-10-07T01:13:41Z)
• REG: A Regularization Optimizer for Robust Training Dynamics [24.850151895583494]
  Row-and-Column-Scaling (RACS) operator regularizes the update steps in a less drastic manner, making it simpler to implement and more compatible with established training dynamics.<n>We demonstrate that our REG achieves superior performance and stability over AdamW, but also maintains consistency with the AdamW training paradigm.
  arXiv  Detail & Related papers  (2025-10-04T06:05:57Z)
• AdaGrad Meets Muon: Adaptive Stepsizes for Orthogonal Updates [5.049533819651459]
  We propose a new adaptive update, AdaGO, which combines a norm-based update with aGrad-type step.<n>AdaGO preserves the orthogonality of the update, which can be interpreted as a spectral descent, while adapting the stepsizes to the optimization landscape by scaling the direction with accumulated past gradients.
  arXiv  Detail & Related papers  (2025-09-03T03:42:22Z)
• Training Deep Learning Models with Norm-Constrained LMOs [56.00317694850397]
  We propose a new family of algorithms that uses the linear minimization oracle (LMO) to adapt to the geometry of the problem.<n>We demonstrate significant speedups on nanoGPT training using our algorithm, Scion, without any reliance on Adam.
  arXiv  Detail & Related papers  (2025-02-11T13:10:34Z)
• MARS: Unleashing the Power of Variance Reduction for Training Large Models [56.67982828148859]
  We propose a unified training framework for deep neural networks.<n>We introduce three instances of MARS that leverage preconditioned gradient optimization.<n>Results indicate that the implementation of MARS consistently outperforms Adam.
  arXiv  Detail & Related papers  (2024-11-15T18:57:39Z)
• Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding [55.107555305760954]
  We propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. Our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.
  arXiv  Detail & Related papers  (2024-01-17T09:01:29Z)
• TWINS: A Fine-Tuning Framework for Improved Transferability of Adversarial Robustness and Generalization [89.54947228958494]
  This paper focuses on the fine-tuning of an adversarially pre-trained model in various classification tasks. We propose a novel statistics-based approach, Two-WIng NormliSation (TWINS) fine-tuning framework. TWINS is shown to be effective on a wide range of image classification datasets in terms of both generalization and robustness.
  arXiv  Detail & Related papers  (2023-03-20T14:12:55Z)
• Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality [65.67315418971688]
  Nearest Orthogonal Gradient (NOG) and Optimal Learning Rate (OLR) are proposed. Experiments on visual recognition demonstrate that our methods can simultaneously improve the covariance conditioning and generalization.
  arXiv  Detail & Related papers  (2022-07-05T15:39:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.