Dead-Direction Conditioners provide gauge-equivariant preconditioning by conditioning optimizer state on symmetry orbits, yielding improved resistance to over-training collapse and higher detection of dead directions compared to AdamW and Muon.
hub
Modular duality in deep learning.arXiv preprint arXiv:2410.21265
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
hub tools
citation-role summary
background 4
citation-polarity summary
roles
background 4representative citing papers
Muon outperforms Adam by reducing curvature penalty via lower Normalized Directional Sharpness, as shown via Taylor approximation on LLM training and proven on stylized quadratic problems with heterogeneous curvature.
The authors derive a Maximally Scale-Stable Parameterization (MSSP) for MoE models that achieves robust learning-rate transfer and monotonic performance gains with scale across co-scaling regimes of width, experts, and sparsity.
Intrinsic Muon provides closed-form linear maximization oracles on multiple Riemannian matrix manifolds for unitarily invariant norms, with convergence rates depending only on manifold dimension or rank.
A unified stochastic convergence theory is developed for adaptive preconditioned first-order methods including AdaGrad variants, Shampoo, and Muon in nonconvex optimization.
Scion is a new stochastic LMO-based optimizer family that unifies existing methods, supports unconstrained problems, and delivers hyperparameter transferability plus speedups on nanoGPT training.
Introduces a class of asynchronous adaptive first-order methods and establishes O(1/sqrt t) convergence (up to logs) for non-convex stochastic optimization under reasonable assumptions.
Muon does not converge on convex Lipschitz functions regardless of learning rate, while error feedback restores theoretical convergence but degrades performance on CIFAR-10 and nanoGPT tasks.
Muon-OGD introduces a spectral-norm constrained orthogonal projection method solved via dual iterations and Newton-Schulz approximations to improve stability-plasticity trade-off in sequential LLM adaptation.
OrScale adds a Frobenius-norm trust-ratio layer-wise scaler to Muon’s orthogonalized updates, with per-layer calibration for language models, yielding higher CIFAR-10 accuracy and better language-model pre-training loss than Muon+Moonlight and AdamW.
Orth-Dion uses QR factorization on the right factor instead of column normalization to eliminate the geometric mismatch in low-rank approximations of spectral optimizers like Muon, achieving O(sqrt(L_r/T)) rate under non-Euclidean smoothness.
Pro-KLShampoo projects KL-Shampoo preconditioners to a spike-and-flat parametric form on an r-dimensional subspace and recovers the full algebraic preconditioner via orthogonalization, outperforming KL-Shampoo on GPT-2 and LLaMA pre-training scales.
Preconditioned matrix norms unify steepest descent, quasi-Newton, and adaptive optimizers, revealing SGD, Adam, Muon, KL-Shampoo, SOAP, and SPlus as special cases and enabling new methods MuAdam and MuAdam-SANIA that are competitive in experiments.
A factored, low-memory optimizer for token embedding and LM-head tables matches Adam's quality while using kilobytes of optimizer state, and token trajectories appear nearly one-dimensional.
Pion is an optimizer that preserves the singular values of weight matrices in LLM training by applying orthogonal equivalence transformations.
HTMuon modifies Muon to produce heavier-tailed updates and weight spectra via HT-SR theory, yielding up to 0.98 lower perplexity on LLaMA pretraining and serving as a plug-in for other Muon variants.
Argues that parametric attention forms are necessary for lifelong in-context learning in transformers to maintain constant memory footprint over arbitrary sequence lengths.
citing papers explorer
-
Dead-Direction Conditioners: Gauge-Equivariant Preconditioning for Deep Networks
Dead-Direction Conditioners provide gauge-equivariant preconditioning by conditioning optimizer state on symmetry orbits, yielding improved resistance to over-training collapse and higher detection of dead directions compared to AdamW and Muon.
-
Why Muon Outperforms Adam: A Curvature Perspective
Muon outperforms Adam by reducing curvature penalty via lower Normalized Directional Sharpness, as shown via Taylor approximation on LLM training and proven on stylized quadratic problems with heterogeneous curvature.
-
How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization
The authors derive a Maximally Scale-Stable Parameterization (MSSP) for MoE models that achieves robust learning-rate transfer and monotonic performance gains with scale across co-scaling regimes of width, experts, and sparsity.
-
Intrinsic Muon: Spectral Optimization on Riemannian Matrix Manifolds
Intrinsic Muon provides closed-form linear maximization oracles on multiple Riemannian matrix manifolds for unitarily invariant norms, with convergence rates depending only on manifold dimension or rank.
-
A unified convergence theory for adaptive first-order methods in the nonconvex case, including AdaNorm, full and diagonal AdaGrad, Shampoo and Muo
A unified stochastic convergence theory is developed for adaptive preconditioned first-order methods including AdaGrad variants, Shampoo, and Muon in nonconvex optimization.
-
Training Deep Learning Models with Norm-Constrained LMOs
Scion is a new stochastic LMO-based optimizer family that unifies existing methods, supports unconstrained problems, and delivers hyperparameter transferability plus speedups on nanoGPT training.
-
Stochastic convergence of parallel asynchronous adaptive first-order methods
Introduces a class of asynchronous adaptive first-order methods and establishes O(1/sqrt t) convergence (up to logs) for non-convex stochastic optimization under reasonable assumptions.
-
Muon Does Not Converge on Convex Lipschitz Functions
Muon does not converge on convex Lipschitz functions regardless of learning rate, while error feedback restores theoretical convergence but degrades performance on CIFAR-10 and nanoGPT tasks.
-
Muon-OGD: Muon-based Spectral Orthogonal Gradient Projection for LLM Continual Learning
Muon-OGD introduces a spectral-norm constrained orthogonal projection method solved via dual iterations and Newton-Schulz approximations to improve stability-plasticity trade-off in sequential LLM adaptation.
-
OrScale: Orthogonalised Optimization with Layer-Wise Trust-Ratio Scaling
OrScale adds a Frobenius-norm trust-ratio layer-wise scaler to Muon’s orthogonalized updates, with per-layer calibration for language models, yielding higher CIFAR-10 accuracy and better language-model pre-training loss than Muon+Moonlight and AdamW.
-
Orth-Dion: Eliminating Geometric Mismatch in Distributed Low-Rank Spectral Optimization
Orth-Dion uses QR factorization on the right factor instead of column normalization to eliminate the geometric mismatch in low-rank approximations of spectral optimizers like Muon, achieving O(sqrt(L_r/T)) rate under non-Euclidean smoothness.
-
Pro-KLShampoo: Projected KL-Shampoo with Whitening Recovered by Orthogonalization
Pro-KLShampoo projects KL-Shampoo preconditioners to a spike-and-flat parametric form on an r-dimensional subspace and recovers the full algebraic preconditioner via orthogonalization, outperforming KL-Shampoo on GPT-2 and LLaMA pre-training scales.
-
Preconditioned Norms: A Unified Framework for Steepest Descent, Quasi-Newton and Adaptive Methods
Preconditioned matrix norms unify steepest descent, quasi-Newton, and adaptive optimizers, revealing SGD, Adam, Muon, KL-Shampoo, SOAP, and SPlus as special cases and enabling new methods MuAdam and MuAdam-SANIA that are competitive in experiments.
-
Token Geometry
A factored, low-memory optimizer for token embedding and LM-head tables matches Adam's quality while using kilobytes of optimizer state, and token trajectories appear nearly one-dimensional.
-
Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation
Pion is an optimizer that preserves the singular values of weight matrices in LLM training by applying orthogonal equivalence transformations.
-
HTMuon: Improving Muon via Heavy-Tailed Spectral Correction
HTMuon modifies Muon to produce heavier-tailed updates and weight spectra via HT-SR theory, yielding up to 0.98 lower perplexity on LLaMA pretraining and serving as a plug-in for other Muon variants.
-
Lifelong In-Context Learning with Transformers Requires Parametric Forms of Attention
Argues that parametric attention forms are necessary for lifelong in-context learning in transformers to maintain constant memory footprint over arbitrary sequence lengths.