Neural Network Training Dynamics

Follow this topic Ask about this topic

Momentum

41 papers in the last four weeks, up 273% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 349

Jun 28, 2026cs.LG

Anti-Collapse Dynamics and the Emergence of Multi-Time-Scale Learning in Recurrent Neural Networks

Long-range learning is hard for recurrent networks trained with stochastic gradient descent, because the influence of a past input fades with the lag

\ell

, and if it fades too fast the dependence cannot be learned from finite data. This fade is captured by an envelope

f(\ell)

. An exponential fade makes the data needed to learn a lag-

\ell

dependence grow exponentially, putting long horizons out of reach; a power-law fade keeps the cost polynomial. We show that the asymptotic decay class of

f(\ell)

is not fixed by the architecture. Instead, it emerges from the coupling between the state dynamics and parameter dynamics, settling into either a collapsed regime (fast, exponential forgetting) or an extended, anti-collapsed regime (slow, power-law forgetting). The intuition is a competition within these coupled dynamics. Training drives the network's effective time scales toward short ones, while rare, heavy-tailed fluctuations of the learning dynamics push a few of them to very long values. The extended regime survives only when these heavy-tailed pushes are strong enough to balance the pull. We make this mathematically precise with a coarse-grained stochastic process and prove exactly when the extended regime exists. A single exponent, the spectral exponent~

β

, then governs both the spread of time scales and how slowly the network forgets. Realizing the regime in practice needs one more ingredient: the joint action of the architecture and the optimizer must be able to hold such a broad spread. A network whose capacity to generate broad time-scale spectra is severely constrained still collapses, even when supplied with strong heavy-tailed forcing. Heavy-tailed fluctuations thus act not as noise to be suppressed, but as the mechanism that sustains long-range learning.

Lorenzo Livi

OPIT – Open Institute of Technology

Heavy-Tailed Noise Recurrent Neural Networks Neural Network Training Dynamics Sequence Modeling

Jun 28, 2026quant-ph

A Coherence Law for Trainability in Noisy Equivariant Quantum Neural Networks

Symmetry provides a quantum neural network structure, but on its own it does not keep the network trainable once noise is present. We ask which physical quantity decides whether the gradients of an equivariant circuit survive decoherence, and we answer with a compact training law. Working with U(1)-equivariant brickwork circuits that conserve a charge, we find that two distinct effects govern a trainable gradient. Causality fixes where the gradient can live, confining it to the backward light cone of the readout inside the active charge sector. Coherence then determines how fast it decays through the contraction of the off-diagonal sector modes that the projected readout can actually observe. We prove a light-cone reduction that pins the noiseless gradient to the sector-restricted cone with a lower bound independent of the total qubit number, and we define a readout-visible aligned coherence rate as a Rayleigh quotient of the noise generator along the gradient-carrying mode. A perturbative open-system analysis turns this rate into a leading-order training law. Density-matrix simulations then confirm that the finite-noise degradation follows a single accumulated variable built from noise depth and coherence contraction, with a coefficient of determination of 0.979. The sharpest test comes from a correlated-dephasing channel that has a large worst-case rate but a near-zero aligned rate. The law predicts no gradient loss for this channel, and none is seen. Sector coherence outperforms every standard channel diagnostic we compare it against, and the analysis identifies readout-visible sector coherence as the quantity that links equivariant architecture, open-system dynamics and noisy trainability.

Hassan Ugail, Newton Howard

Centre for Visual Computing and Intelligent Systems University of Bradford United Kingdom · School of Individualized Study Rochester Institute of Technology United States

Equivariant Neural Networks Neural Network Training Dynamics Quantum Circuit Trainability Quantum Neural Networks

Jun 28, 2026cs.LG

On the Nonlinearity of Learning Rate Scaling for LLM Training

Learning-rate transfer can reduce the cost of training large language models: instead of sweeping learning rates at target scale, practitioners extrapolate from smaller runs. Existing approaches often assume that the optimal learning rate follows a log-linear scaling law in data scale and model size. We carefully examine and evaluate this scaling law. In our empirical study of GPT-2--style models from 22M to 707M parameters trained on 5B to 100B tokens, the optimal learning rate develops upward curvature at larger scales, leading to inaccurate extrapolation. We find that this curvature largely disappears when learning rates are replaced by effective learning rate (the step size in normalized weight space), and when data

D

extrapolation is used instead of model size

N

extrapolation. Next, we explain nonlinearity in scaling: weight-norm converges to equilibrium slower when optimal learning is small, requiring a larger step size to reduce the transient phase. Experiments with AdamH, which directly controls the effective learning rate, further support this explanation.

Zaiwen Yang, Huaqing Zhang, Jing Xu +1

Tsinghua University · Tsinghua University, Shanghai Qi Zhi Institute

Language Model Scaling Laws Hyperparameter Transfer Neural Network Training Dynamics LLM Training

Jun 26, 2026cs.LG

A Gravitational Interpretation of Safety Reversion under Fine-Tuning

Safety alignment in large language models can degrade during post-training even when neither the data nor the objective is intentionally adversarial. Alignment rebound and reverse dynamics suggest that this degradation may reactivate behavior suppressed during safety alignment. Building on these ideas, we hypothesize that ordinary non-adversarial post-training follows a reversion direction: the activation-space displacement from the safety-aligned model toward a more permissive, earlier helpful-only state. We see that for Llama, every tested trajectory across references, tasks, and seeds exceeds a matched empirical null, while at aligned Llama and Qwen checkpoints, a vocabulary readout shows that the direction locally favors task-engaging over fixed refusal-like openings. Its geometric expression is behaviorally informative: as post-training proceeds, alignment with the direction and harmfulness increase together, yielding a strong descriptive correlation (Spearman r=0.958). To move beyond correlation, we test causal relevance during adaptation using objectives constructed from this coordinate. Across all tested Llama, Qwen, and Gemma settings from 3B to 14B, an optimizer-matched objective opposing positive motion reduces geometric alignment and harmfulness relative to ordinary fine-tuning, whereas a separately stabilized objective reinforcing that motion increases both. Every model and scale exhibits the same mean block-baseline-push ordering, showing that the causal relevance of the reversion direction is not tied to one architecture or model size. Finally, we show that a standard safety-rehearsal objective, built without access to the direction, independently opposes it and cuts cumulative reversion by about 30% in Llama and Qwen.

Samuele Poppi, Nils Lukas

Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), UAE

Fine-Tuning LLM Safety Alignment Neural Network Training Dynamics

Jun 26, 2026cond-mat.dis-nn

Spectral phase transitions and trainability in neural network learning dynamics

The emergence of low-dimensional structures in the spectra of neural network weight matrices is a common empirical feature of trained models, but the dynamical origin of this phenomenon during learning remains an open problem. We formulate neural network training as the stochastic evolution of an initially random matrix ensemble, driven by stochastic gradient descent (SGD) updates that reshape the spectral bulk while amplifying signal strength. This induces a Baik-Ben Arous-Péché (BBP) transition during training, where isolated eigenvalues detach from the random bulk distribution, providing a dynamical framework for representation formation in high-dimensional learning dynamics. We demonstrate this in a solvable linear teacher-student model, where spectral evolution is analytically tractable and a phase diagram of trainability governed by the step size (or learning rate) and initial weight variance is obtained, and subsequently extend our formalism beyond the linear regime to nonlinear and stochastic settings. Numerical simulations in realistic settings support this picture, showing robust emergence of spectral alignment during training. Our results suggest that spectral analysis may provide a unified perspective of stochastic learning dynamics, linking trainability, optimisation hyperparameters, spectral phase transitions, and representation learning in neural networks.

Chanju Park, Dario Bocchi, Francesco D'Amico +2

Centre for Quantum Fields and Gravity, Department of Physics, Swansea University, Swansea SA2 8PP, United Kingdom · Physics Department, University of Rome Sapienza, Piazzale Aldo Moro 5, 00185 Rome, Italy · Institute of Nanotechnology, CNR-NANOTEC, Piazzale Aldo Moro 5, 00185 Rome, Italy +1

Random Matrix Theory Neural Network Training Dynamics Stochastic Gradient Descent

Jun 26, 2026cs.LG

Singular Learning and Occam's Razor in Deep Monomial Networks

In the optimization of neural networks, gradient dynamics are influenced by critical points that arise from the model's architecture. These critical points occur where the Jacobian of the model's parametrization is rank-deficient, and are the most pronounced singularities studied in Singular Learning Theory. We investigate such points in deep fully-connected networks with monomial activations via tools from polynomial algebra such as Mason's Theorem. We show that, for sufficiently large activation degree, criticality occurs precisely at subnetworks, i.e., at parameter configurations where some neurons are inactive or redundant. This offers a mathematical perspective on the implicit bias in deep neural networks, explaining the tendency of these models to converge toward simpler functions.

Kathlén Kohn, Giovanni Luca Marchetti, Farhan Shabir +2

KTH Stockholm · No Affiliation · Umeå University +1

Neural Network Optimization Singular Learning Theory Neural Network Training Dynamics Statistical Learning Theory

Jun 26, 2026cs.LG

Layerwise Progressive Freezing: A Training Scaffold for Depth-Scalable Binary Networks

Training binary neural networks (BNNs) from scratch is dominated by the straight-through estimator (STE), whose forward/backward mismatch produces severe accuracy degradation as networks deepen. We study an orthogonal axis: when and where binarization is enforced during training. We introduce StoMPP (Stochastic Masked Partial Progressive Binarization), which gradually replaces clipped weights and activations with their hard binary counterparts layer by layer from input to output, using stochastic partial masks with soft refresh. StoMPP delivers two complementary benefits. As a standalone training rule, it provides a fully STE-free procedure that improves over vanilla STE with gains that grow with depth (ResNet-50 BNN: +18.0/+13.5/+3.8 on CIFAR-10/100/ImageNet), and the pattern holds across ResNet-18/34/50, MobileNetV2, and BERT fine-tuning. Composed with surrogate gradients by applying STE only to frozen entries, it reaches +27.1/+19.8/+17.7 over vanilla STE on the same setting. Underlying both regimes is a single mechanistic finding: progression order is decisive. Forward layerwise progression prevents depth collapse, reverse progression collapses to near-chance, and binary-weight networks (without binary activations) are insensitive to order. We trace this asymmetry to activation-induced gradient blockades: a committed binary activation severs gradient flow upstream, and ordering controls when these blockades form. To isolate the progression's contribution from any benefit conferred by STE, we conduct all ablations in the STE-free regime; the resulting characterization (schedule, refresh, ordering, dynamics) thus reflects the progression itself rather than its interaction with surrogate gradients.

Evan Gibson Smith, Bashima Islam

Worcester Polytechnic Institute

Binary Neural Networks Activation Quantization Neural Network Training Dynamics

Jun 26, 2026cs.LG

Aurora: A Leverage-Aware Spectral Optimizer

We show that for tall matrix parameters, like projection matrices in the MLP layers, the Muon update can have row norms that are arbitrarily non-uniform. This can lead to a self-reinforcing feedback loop whereby neurons receive persistently small updates and eventually do not contribute meaningfully to network outputs. This problem is effectively mitigated by an additional row normalization step, but current methods do this in a way that moves the Muon update geometry away from the polar factor of the momentum matrix, which we find is undesirable. We propose Aurora, an optimizer that enforces row-uniformity of matrix parameter updates while respecting Muon's polar factor geometry. Aurora outperforms Muon in our pre-training experiments and, when combined with existing methods, achieves state-of-the-art performance among spectral optimizers on the optimizer track of the modded-nanoGPT speedrun. Additionally, we find that Aurora's empirical gains over Muon scale with the MLP expansion factor, suggesting that Aurora may allow for effective training of very wide MLP layers.

Alec Dewulf, Dhruv Pai, Li Yang +2

Tilde Research

Neural Network Optimization Muon Optimizer Matrix Optimization Spectral Methods Neural Network Training Dynamics

Jun 25, 2026cs.LG

Effective Covariance Dynamics in Solvable High-Dimensional GANs

We study a solvable high-dimensional model of generative adversarial network (GAN) training in which a linear generator learns a low-dimensional subspace from data with structured latent covariance. Prior solvable GAN analyses assume unconditional signals with diagonal latent covariance; we extend the multi-feature discriminator setting to class-dependent, correlated, and non-zero-mean latent structure. For the quadratic energy discriminator, all such heterogeneity enters the dynamics through a probability-weighted effective second moment. We prove that the stochastic microscopic training process converges, in the high-dimensional limit, to deterministic ordinary differential equations governed by this effective covariance. In the matched-covariance specialization, the stability analysis yields a mode-wise solvable interval determined by the learning rates and noise level: learning begins when the leading effective eigenvalue crosses the lower threshold, while full recovery requires all relevant effective modes to remain within the interval. This reveals a signal-boosting mechanism: low-rank correlations can lift weak directions above the learnability threshold, whereas overly strong correlations destabilize recovery. Numerical simulations validate the ODE, phase boundary, and boosting mechanism. Experiments on MNIST, FashionMNIST, and CIFAR-10 further show that informed generator covariance improves alignment with the data-driven reference subspace.

Andrew Bond, Zafer Doğan

MLIP Group, KUIS AI Center, Koç University, İstanbul, Turkey. · Dept. of EE Engineering, Koç University, İstanbul, Turkey.

Neural Network Training Dynamics Generative Adversarial Networks

Jun 24, 2026cs.LG

Improving Neural Network Training by Decoupling the Magnitude and Direction of Weight Vectors

Modern neural network training relies on optimizers such as Adam and Muon which act on each weight matrix as a single object. Yet every weight matrix carries two distinct quantities -- a \emph{magnitude} and a \emph{direction} -- and all optimizers stepping in the matrix as a whole couple their dynamics: the directional change from an update depends on the current magnitude, while the magnitude drifts as a byproduct of learning the direction, so neither is governed directly by the learning rate. Typical training therefore leans on surrounding recipes such as weight decay and warmup to keep learning stable at scale, though these regulate the coupling only indirectly; other recent methods instead constrain the weight to a fixed-norm sphere, but add no learnable magnitude, leaving scale control to normalization layers alone. We propose \emph{Magnitude--Direction (MD) Decoupling}, an optimizer modification that 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, all while the model still sees a single fused weight tensor. The method is agnostic to the base optimizer and removes the need for weight decay and warmup. 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 (MoE) models. Treating magnitude and direction as separately controlled quantities thus yields more predictable training dynamics and a simple, broadly applicable improvement to modern optimizers.

Alexander Hägele, Alejandro Hernández-Cano, Atli Kosson +1

Machine Learning and Optimization Lab, EPFL

Neural Network Optimization Neural Network Training Dynamics

Jun 23, 2026cs.LG

Training Dynamics of Neural Software Defect Predictors under Coupled Data-Quality Issues

Context: Software defect prediction supports maintenance decisions such as testing prioritization, release-risk assessment, and quality monitoring. However, metric-based SDP datasets often contain coupled data-quality issues, especially class imbalance and class overlap. Prior work has mainly measured their impact through endpoint performance, while recent evidence suggests that such issues may also appear in neural training dynamics (gradients, weights, biases, error trajectories). However, these studies examine issues in isolation, leaving open how internal neural network training patterns manifest when data quality issues are coupled. Objective: We investigate how training-dynamics patterns from class imbalance, overlap, and their coupling can be characterized under interaction-aware conditions in deep learning-based SDP. Method: We conduct a controlled intervention study on class-level UBD datasets, training a fixed MLP under imbalance-only, overlap-only, and joint conditions across five seeds. Training dynamics are logged per epoch; fidelity is monitored via coupling ratios. Patterns are characterized using effect sizes, trajectories, sensitivity analyses, and rule-based classification. Expected contribution: The study will produce an interaction-aware empirical protocol and a candidate taxonomy of training-dynamics patterns for coupled data-quality issues in metric-based SDP.

Emmanuel Charleson Dapaah, Philip Makedonski, Jens Grabowski

Institute of Computer Science University of Goettingen Goettingen, Germany

Class-Imbalanced Learning Neural Network Training Dynamics

Jun 23, 2026cs.LG

How Complexity Contributes to Learning Opacity in Machine Learning

Machine learning (ML) algorithms are known to be opaque. We do not know the reasons for their predictions. The learning process leading to the prediction function is also opaque. We do not fully understand the time evolution of the weight values of neural nets (NN) and related dynamical phenomena. While prediction opacity is widely studied, learning opacity remains largely underexplored. This article studies learning opacity trough the lens of complex dynamical systems. We argue that NN learning is essentially a complex system and that learning opacity is due to dynamical complexity and the epistemological challenges that arise from it. We identify three key properties of training complexity -- sensitivity to weight initialization, feedback in gradient based optimization, and sensitivity to the training data -- and show how each contributes to learning opacity. As these properties are fundamental to the learning process damping or eliminating them would fundamentally alter how ML systems learn. Some sources of opacity in ML may hence be irreducible.

Joachim Stein, Eric Raidl

1Heidelberger Akademie der Wissenschaften · 2Universität Tübingen, Cluster of Excellence: Machine Learning – New

Neural Network Training Dynamics Gradient Descent Dynamics

Jun 23, 2026cs.LG

When Top-1 Fails: Calibrating LoRA Monitors for Masked Diffusion LMs

Discrete diffusion language model (DLM) fine-tuning inherits inexpensive diagnostics from denoising-time confidence monitors, but their PEFT-training meaning is untested. We test top-1 argmax concentration as a collapse warning. Across 816 LoRA/PEFT configurations from three DLM families, the warning fires for every configuration while logs record 0/816 actual collapses at the 200 step horizon, giving zero precision. The cause is pre-equilibrium saturation: top-1 concentration is already high before optimization and quickly becomes insensitive to final training stability. We then evaluate max LoRA gradient norm, a parameter-side signal that samples gradient routing rather than token concentration. On a pooled held-out LLaDA-family split, a train-optimized threshold identifies top-decile final-loss configurations with precision 0.68 and F1=0.79, above the all-positive top-1 baseline even at the lower split-bootstrap confidence bound. Autoregressive controls and cross-family threshold failures bound the result to short-horizon DLM-LoRA inspection rather than a universal collapse detector. Workflow: drop top-1 as a PEFT alarm, log max-gradient early in training, and calibrate thresholds per DLM family before routing runs for inspection.

Lucky Verma, Pratik Yadav

Independent Researcher · University of Maryland, Baltimore County

Diffusion Model Fine-Tuning Diffusion Language Models Neural Network Training Dynamics Parameter-Efficient Fine-Tuning

Jun 22, 2026cs.LG

Convergence of Gradient Descent for General Neural Network Architectures Beyond the NTK Regime

Training dynamics is central to understanding neural networks, yet its theoretical analysis remains difficult even for simple architectures and becomes substantially more challenging for general modern architectures. In this paper, we propose a convergence framework for analyzing gradient descent (GD) dynamics under a broad family of neural network architectures and datasets beyond the neural tangent kernel (NTK) regime. The framework is formulated at the level of network blocks and covers architectures including pre-normalized multi-layer transformers. More precisely, under mild assumptions, we prove that for almost all initializations, GD with regular learning rates converges to the neighbourhood of a stationary point. This is mainly proved by establishing an iterate-dependent PL-type inequality through analyticity and measure-zero arguments, and by proving Lipschitz smoothness along the GD trajectory through polynomial generalized smoothness and a local relaxed dissipative condition. We further interpret the theorem under Xavier initialization and practical architectural scaling, showing that the learning rate scale depends on the depth and effective bottleneck dimensions rather than the largest width. Finally, we derive structural nondegeneracy implications for residual connections and function composition, and provide a generic characterization of global minimizers within our framework.

Yuqing Wang

Johns Hopkins University

Gradient Descent Neural Network Optimization Neural Network Training Dynamics

Jun 22, 2026cs.LG

GRAIN: Group Aggregation via Min-Norm Objective

Learning instability is a long-standing problem across machine learning, but it is especially acute in the overparameterized regime that defines modern deep learning: large models fine-tuned or trained on limited data traverse flat loss landscapes with many nearly-equivalent minima, and stochastic factors (initialization, data order, dropout, hardware non-determinism) can route optimization to very different solutions. The rise of large pretrained models (LPMs) makes the problem more urgent: training cost is high, downstream data is often small, and repeated runs for variance reduction are prohibitive. We introduce \textbf{GRAIN} (\textbf{G}roup \textbf{A}ggregation via m\textbf{IN}-norm objective), a lightweight training algorithm that replaces the mean aggregation used in mini-batch optimization (both across mini-batches and within a mini-batch) with a min-norm convex combination of group-wise gradients. \mName guarantees a non-negative inner product between the aggregated update and every group gradient, resolving intra- and inner-batch gradient conflict, and retains an

\mathcal{O}(1/T)

convergence rate comparable to SGD. Under mild smoothness and absolute-continuity assumptions, the min-norm solution differs almost surely from the arithmetic mean, which yields a uniform-stability bound for \mName strictly tighter than the standard bound for SGD. Empirically across generation, classification, and regression at LPM scale, \mName delivers consistent improvements in mean performance and reductions in run-to-run variance over a broad suite of tasks, with no extra training-time or storage cost beyond a single backward pass.

Nghia Bui, Jiarui Yao, Lijing Wang

Department of Data Science New Jersey Institute of Technology · Boston Children’s Hospital Harvard Medical School

Algorithmic Stability Neural Network Training Dynamics Stochastic Gradient Descent

Jun 21, 2026cs.LG

Noise-Debiased Thermodynamic Variance for Local Learning Coefficient Probes

Local learning coefficient (LLC) probes offer a singularity-aware view of neural-network training, but mean-energy methods require a local loss baseline that is ambiguous at transient checkpoints. Thermodynamic variance avoids this input; under mini-batch evaluation, however, direct variance mixes cross-state loss fluctuations with same-state noise. We operationalize this route with the \emph{Shift-Invariant Variance Estimator} (SIVE), which estimates and subtracts the latter component using repeated evaluations. Conditional on any fixed retained path, unclipped SIVE is unbiased for noiseless path variance without requiring MCMC stationarity. The finite-scale diagnostic remains indexed by localization scale

h

---even a locally linear loss has tether-dependent variance---while interpretation as a Real Log Canonical Threshold (RLCT) requires additional stationary low-temperature conditions. Toy experiments recover calibrated finite-scale targets. At the primary localization scale, all five MNIST MLP trajectories exhibit a mid-training trough followed by a rebound in SIVE, while Raw Variance decreases from Epoch 40 to 100 in every trajectory. Across four localization scales, the joint early-drop/late-rise criterion is met in 19 of 20 trajectory--scale pairs. At Epoch 40, the estimated observation-noise correction accounts for

77.5\%

of Raw Variance. Same-state debiasing thus reveals a reproducible turning structure masked by time-varying observation noise.

Yingjia Cai

Singular Learning Theory Neural Network Training Dynamics Statistical Learning Theory Local Learning

Jun 20, 2026cs.LG

Mitigating Early Training Collapse in CTR Models

Deep neural models for click-through rate prediction often exhibit a sharp decline in validation performance immediately after the first training epoch despite continued improvement in training loss. This instability restricts effective learning and limits model performance. In this study, we analyze this behavior using large-scale industrial datasets and evaluate practical mitigation strategies. While reducing the learning rate provides only incremental gains, controlling feature sparsity yields substantial improvements. Removing highly sparse features and aggregating infrequent feature values stabilizes training, extends useful learning beyond a single epoch, and improves both offline evaluation metrics and online system performance.

Ergun Biçici, Erkan Çetinyamaç

Intelligent Application Development, Huawei Türkiye R&D Center, Istanbul, Turkey

CTR Prediction Feature Selection Neural Network Training Dynamics User Preference Modeling

Jun 19, 2026cs.LG

Expressivity Saturation: Reduced Affine Region Usage Under Increasing Task Complexity

Piecewise-affine neural networks (e.g., with ReLU or LeakyReLU activations) implement continuous piecewise-affine maps, and the number of affine regions provides a natural proxy for expressive capacity. However, the gap between theoretical region capacity and the affine regions realized after training remains insufficiently understood. We study this gap from two complementary perspectives. First, we give a rigorous, architecture-dependent theorem for affine line-segment probes: for multilayer perceptrons with piecewise-affine activations, the number of affine pieces realized along an affine line-segment probe is upper bounded by an explicit product of layer-wise width terms (and activation breakpoint factors). This yields a neuron-threshold lower bound for representing target functions with prescribed one-dimensional piece complexity, formalizing the minimal region budget required for complex signals. Second, we exactly enumerate affine regions realized within bounded 2D and higher-dimensional domains under controlled task complexity. Under fixed architectures and training protocols, increasing input--label complexity yields trained solutions with markedly fewer realized regions in the evaluation domain, even though worst-case architectural capacity is unchanged; we call this reduced region usage expressivity saturation. Moreover, in the most challenging regimes, 2D visualizations show that region-usage collapse often coincides with degraded decision boundaries. Finally, we visualize the training dynamics of affine-region partitions and decision boundaries, revealing a consistent refinement process during optimization.

Xuan Qi, Yi Wei, Fanqi Yu +1

AI for Good (AIGO), Istituto Italiano di Tecnologia · DITEN, University of Genoa · State Key Laboratory of Novel Software Technology +1

Multilayer Perceptrons Neural Network Approximation Theory Neural Network Training Dynamics Neural Network Expressivity

Jun 18, 2026stat.ML

Statistical Properties of Training & Generalization

Deep learning has managed to evade numerous intuitions from classical statistics to achieve unprecedented performance on a number of real-world tasks. In this article, we investigate the key features and surprises of deep learning from a physics-informed perspective, taking care to point out and justify where possible the many choices inherent in constructing a deep learning model. In particular, we review the phenomenon of neural scaling laws and discuss their interplay with the constraints and inductive biases which may be present when applying machine learning to problems in physics.

Itay Lavie, Noam Levi, Yonatan Kahn

John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA · Tel Aviv University, Tel Aviv, Israel · Department of Physics, University of Toronto and Vector Institute, Toronto, ON, Canada

Neural Network Generalization Scientific ML Neural Network Training Dynamics Neural Scaling Laws

Jun 18, 2026cs.LG

Physics-Informed Neural Network with Squeeze-Excitation-like Attention

We introduce SEA-PINN, a novel architecture that incorporates a Squeeze-Excitation-like attention mechanism into physics-informed neural networks to dynamically recalibrate the importance of neurons across layers. A key feature of SEA-PINN is its highly stable initialization. On 17 out of 20 benchmark problems, SEA-PINN exhibit nearly negligible variance and significantly reduced initial loss, establishing a quasi-deterministic and favorable starting point for optimization. Notably, without employing Fourier feature embeddings or periodic activation functions, SEA-PINN attained competitive accuracy (83% vs. 90% improvement relative to FNN-PINN on the high-frequency case 7) as compared with TSA-PINN-a model specifically engineered for high-frequency problems via learnable frequencies in sinusoidal activations. Furthermore, integrating SEA-PINN into TSA-PINN boosted performance by 42.49%. These results underscore SEA-PINN as a lightweight plug-in module that enhances nonlinear representation power, promotes more robust and efficient convergence, and strengthens the overall reliability of physics-informed learning.

Yun-Fei Song, Long-Gang Pang, Fu-Peng Li +1

Key Laboratory of Quark and Lepton Physics (MOE) & Institute of Particle Physics, Central China Normal University, Wuhan 430079, China · Artificial Intelligence and Computational Physics Research Center, Central China Normal University, Wuhan 430079, China · Key Laboratory of Nuclear Physics and Ion-beam Application (MOE) & Institute of Modern Physics, Fudan University, Shanghai 200433, China +2

Neural PDE Solvers Physics-Informed ML Neural Network Training Dynamics

Jun 16, 2026cs.LG

What Does the Weight Norm Control in Grokking? Logit-Scale Mediation under Cross-Entropy

Grokking, the delayed jump from memorization to generalization, is usually tied to the weight norm: a smaller norm generalizes sooner. We ask what the norm actually controls. Holding the weight norm fixed by clamping and varying only an output temperature, we slide the grokking delay across its entire norm-induced range under cross-entropy; matching the effective logit scale back to baseline recovers about 85% of the delay at two moduli. Across a grid of norms and temperatures the delay collapses onto the logit scale alone (R2 = 0.97), with the norm adding 1-2% beyond it. The effect is loss-dependent: under mean-squared error the logit scale is pinned and the norm acts through a different route. A memorization control, a float64 softmax-collapse audit, and a no-LayerNorm transformer point to the same channel. Forking arms from one identical state, the delay follows the held norm value and not the clamp operation, which closes a rescaling-artifact concern. The proximal variable is the logit scale and the softmax saturation it drives; the weight norm is only an upstream handle. All numbers, tables, and figures reproduce from released code and data.

Truong Xuan Khanh

H&K Research Studio, Clevix LLC · Hanoi, Vietnam

Neural Network Generalization Neural Network Training Dynamics Grokking

Jun 16, 2026cs.LG

Edge Flow: A Tractable and Predictive Continuous-Time Model for Gradient Descent at the Edge of Stability

Gradient descent in deep learning may operate at the edge of stability (EoS), a regime in which the largest eigenvalue of the loss Hessian hovers near the stability threshold

2/η

, where

η

is the learning rate. Classical analysis tools such as gradient flow and the descent lemma do not apply here, motivating the search for a continuous-time model valid at EoS. We propose Edge Flow, a system of three coupled ordinary differential equations that provides a tractable, faithful, and predictive model of gradient descent dynamics at EoS. Edge Flow decomposes the dynamics into a center, an oscillation direction, and an oscillation magnitude. The center follows a modified gradient flow on a symmetrized loss; the direction tracks a top eigenvector of the Hessian via Rayleigh quotient dynamics; and the magnitude grows or decays exponentially depending on whether the sharpness exceeds or falls below the threshold

2/η

. Crucially, sharpness stabilization emerges from the coupled dynamics via a self-stabilization feedback loop. Discretizing Edge Flow only requires two gradient evaluations and one Hessian--vector product at each iteration. We demonstrate empirically that Edge Flow tracks the dynamics of gradient descent at least as faithfully as previously proposed continuous-time EoS models, while in addition resolving the oscillation of the sharpness at the onset of EoS, and that it provides a principled framework for understanding and mitigating instabilities in this regime.

Pierre Marion

Inria, École Normale Supérieure, PSL Research University

Edge of Stability Neural Network Training Dynamics Gradient Descent Dynamics

Jun 16, 2026cs.LG

Conservation Laws for Modern Neural Architectures

Understanding gradient descent dynamics is key to explaining the success of over-parameterized models, where implicit bias manifests through conservation laws in gradient flow. While such laws are well understood for linear and ReLU networks, they remain largely unexplored for modern architectures. This work develops a unified framework to characterize conservation laws for contemporary models, including feedforward networks with GELU, SiLU, and SwiGLU activations, multihead attention with sinusoidal and rotary positional encodings, and Mixture-of-Experts architectures under diverse gating designs. Our theoretical findings are supported by experiments that validate the predicted invariants.

Viet-Hoang Tran, Vinh Khanh Bui, Tan Lai Ngoc +3

National University of Singapore · Center for AI Research, VinUniversity · Independent Researcher +1

Neural Network Training Dynamics Multi-Head Attention Gradient Descent Dynamics

Jun 16, 2026cs.LG

A Link between Shock-wave Theory and Symmetry-reduced Stochastic Gradient Descent for Artificial Neural Networks

We develop a mathematically explicit link between shock-wave theory and the symmetry-quotiented learning dynamics of stochastic gradient descent, drawing on differential geometry, Lie group theory, and fluid mechanics. Specifically, after quotienting parameter symmetries and applying local-entropy coarse-graining, the effective dynamics satisfy a viscous Hamilton--Jacobi equation on the quotient manifold. Moreover, under the assumption that the raw parameter dynamics can be summarized by a gradient field on the quotiented space, the gradient of the coarse-grained loss function obeys a Burgers-type equation, and shock formation can be established rigorously. We apply our theory to multilayer perceptrons, convolutional neural networks, Transformers, and mean-field networks, and show that they obey the Hamilton--Jacobi or Burgers-type equations. We conjecture that this framework also yields practical diagnostics for deep learning. In architectures such as Transformers, raw parameter norms are often distorted by symmetry redundancy and may therefore be misleading, whereas symmetry-corrected quotient observables provide a principled basis for monitoring, forecasting, and controlling training-phase transitions.

Taiki Miyagawa

NEC Corporation

Neural Network Training Dynamics Gradient Descent Dynamics Stochastic Gradient Descent

Jun 15, 2026math.OC

Exploding and vanishing gradients in deep neural networks: the effect of residual connections

The well known phenomenon of exploding and vanishing gradients in deep neural networks is analyzed using multiplicative ergodic theory. The effect of adding a residual connection is explained in this context. Specifically, a characterization of Liapunov exponents due to Furstenberg and Kifer is exploited in order to make a precise statement about the Liapunov spectrum and the effect of residual connections on it.

Vivek S Borkar

Residual Learning Neural Network Training Dynamics

Jun 15, 2026cs.LG

Taming Curvature: Architecture Warm-Up for Stable Transformer Training

Training billion-parameter Transformers is often brittle, with transient loss spikes and divergence that waste compute. Even though the recently developed Edge of Stability (EoS) theory provides a powerful tool to understand and control the stability of optimization methods via the (preconditioned) curvature, these curvature-controlling methods are not popular in large-scale Transformer training due to the complexity of curvature estimation. To this end, we first introduce a fast online estimator of the largest (preconditioned) Hessian eigenvalue (i.e., curvature) based on a warm-started variant for power iteration with Hessian-vector products. We show theoretically, and verify empirically, that the proposed method makes per-iteration curvature tracking feasible at billion parameter scale while being more accurate. Using this tool, we find that training instabilities coincide with surges in preconditioned curvature and that curvature grows with depth. Motivated by these observations, we propose architecture warm-up: progressively growing network depth to carefully control the preconditioned Hessian and stabilize training. Experiments on large Transformers validate that our approach enables efficient curvature tracking and reduces instabilities compared to existing state-of-the-art stabilization techniques without slowing down convergence.

Sameera Ramasinghe, Ajanthan Thalaiyasingam, Hadi Mohaghegh Dolatabadi +6

Pluralis Research

Edge of Stability Transformer Second-Order Optimization Neural Network Training Dynamics

Jun 15, 2026cs.LG

Noise-Driven Escape from Metastable Phases explains Grokking in Deep Neural Networks

Deep neural networks (DNNs) exhibit first order phase transitions under variations of the L2 regularization strength, with each transition marking the onset of a new learnable feature. Below a critical regularization strength, all features are in principle learnable, but coexisting metastable states, separated by energy barriers, can trap the network and impede convergence. A strength of DNNs is their ability to generalize. But many open questions remain, among them the origin of so called grokking: the abrupt, delayed onset of generalization after prolonged apparent overfitting. We show for linear DNNs that grokking is consistent with hysteresis in first-order L2 phase transitions: using L2 regularization to engineer deliberate trapping, we demonstrate that a model in a low-accuracy metastable state escapes only when SGD noise drives it across an energy barrier, with escape times following Arrhenius scaling. We reproduce grokking-like delayed convergence across two orders of magnitude in escape time by deliberately trapping models in metastable phases. Using sparse sub-sampling we also reproduce the canonical grokking curve where test error eventually approaches the final training error. Our work suggests that the number of metastable states equals the number of learnable features -- one per singular value of the data covariance -- the potential for hysteresis grows naturally with task complexity. We provide evidence that the same mechanism likely operates in general nonlinear DNNs. Our results provide routes toward more efficient learning schemes.

Ibrahim Talha Ersoy, Karoline Wiesner

Complexity Science Group, Institute of Physics and Astronomy, University of Potsdam, Potsdam, Germany

Statistical Physics of Learning Neural Network Generalization Phase Transitions Neural Network Training Dynamics Grokking

Jun 14, 2026cs.LG

Z-Plane Neural Networks: Bounded Geometric Activation Replaces ReLU and LayerNorm

Modern deep neural networks rely on Euclidean scalar activations (e.g., ReLU) and global normalization techniques (e.g., LayerNorm) to prevent gradient instability in deep architectures. However, these mechanisms inherently cause dead neurons, discard critical directional information, and destroy the orthogonality of feature representations. Inspired by the frequency-modulation transmission of biological axons, we propose the Z-Plane Neural Network, which maps hidden states into 2D phasor bundles on a hypersphere. We introduce a novel geometric activation function, Radial Bounding(

\mathbf{x} / \max(1, \|\mathbf{x}\|_2)

), which limits the energy magnitude while preserving the phase (direction). We demonstrate mathematically that this isotropic activation maintains 1-Lipschitz continuity and prevents gradient vanishing by preserving tangential gradients. Empirically, a 100-layer Z-Plane Multi-Layer Perceptron (MLP)-entirely devoid of ReLU and LayerNorm-successfully converges on the MNIST dataset with 98.34% accuracy and absolute numerical stability, proving that bounded geometric activation alone is sufficient for stable deep learning.

Sungwoo Goo, Hwi-yeol Yun, Sangkeun Jung

College of Pharmacy, Chungnam National University, Daejeon, Republic of Korea · Department of Computer Science & Engineering, Chungnam National University, Daejeon, Republic of Korea

Neural Representation Geometry Neural Network Optimization Neural Network Activation Functions Neural Network Training Dynamics

Jun 14, 2026cs.LG

A Bifurcation Theory Framework for Gradient Descent on the Edge of Stability

The Edge of Stability (EoS) phenomenon, where gradient descent operates with sharpness exceeding the classical convergence threshold yet the loss decreases over long timescales, is ubiquitous in modern deep learning but remains poorly understood in realistic settings. Prior rigorous analyses have been largely confined to scalar or low-dimensional losses with specific structural forms. In this work, we develop a bifurcation theory framework for gradient descent on the edge of stability that applies directly to overparameterized neural networks. By decomposing the training dynamics into components normal and tangent to the manifold of minimizers, we show that stable EoS training arises from a flip bifurcation in the normal direction, governed by the sign of the first Lyapunov coefficient, while the tangent dynamics drift toward regions of decreasing sharpness. Under mild spectral and geometric assumptions on the loss landscape, we prove convergence to the minimizing manifold when training at the EoS threshold. As a corollary, we recover and unify prior results: we show that the product-stability condition of Gan (2026) is an instance of our framework.

Eric Gan

Independent Researcher.

Dynamical Systems Edge of Stability Neural Network Training Dynamics Gradient Descent Dynamics

Jun 13, 2026math.OC

A Conservation Law for Equilibrium Propagation and Coupled Learning

In this paper we show that the physical learning methods known as coupled learning (CL) and equilibrium propagation (EP) conserve a mass-like quantity in the trainable parameters in the continuous-time, small-nudging limit. We prove that this conservation holds in a broad range of physically relevant settings. We then show that the conservation law constrains the training dynamics in a way that makes convergence reliable in important settings for linear circuits. We conclude by discussing some practical implications of this conservation law.

Joshua A. McGinnis, Adam G. Kline, Yoichiro Mori

Department of Mathematics, University of Pennsylvania, Philadelphia, PA, USA · Department of Physics & Astronomy, University of Pennsylvania, Philadelphia, PA, USA · Department of Biology, University of Pennsylvania, Philadelphia, PA, USA

Equilibrium Propagation Neural Network Training Dynamics