Skip to content
Preprint

Hyperparameter Transfer in Graph Neural Networks

Jul 2026 · 0 citations · 34 references
Computer Science

TL;DR

A transfer parameterization is developed and validated for GNNs trained with SGD, Adam, and AdamW that yields stable feature updates, learning rate transfer, and improved performance as width and depth increase, and provides a practical recipe for scaling GNNs across a variety of learning tasks and training scenarios.

Abstract

The performance of deep learning models crucially depends on the settings of hyperparameters like learning rate, initialization scale, and weight decay. Hyperparameter transfer aims to make near-optimal hyperparameter settings consistent across model scale, so that large models can be optimized by proxy tuning their smaller, cheaper-to-optimize counterparts. While transfer principles are well-studied in the context of dense neural networks in language and vision tasks, they remain comparatively under-explored for graph neural networks (GNNs). We develop and validate a transfer parameterization for GNNs trained with SGD, Adam, and AdamW. Through theoretical scaling analyses and controlled experiments, we show that the proposed parameterization yields stable feature updates, learning rate transfer, and improved performance as width and depth increase. For SGD, we identify graph-dependent first-layer correction factors and show that their use can accelerate early training in graphs with sparse bag-of-words inputs. For Adam, we explore how different message passing normalizations affect early- and late-training transfer behavior, illustrating the importance of message passing normalization and advocating for an associated hyperparameter. For AdamW, we adapt a parameterization that allows for the joint transfer of weight decay and learning rate. Together, these results provide a practical recipe for scaling GNNs across a variety of learning tasks and training scenarios.

View source

Similar papers

Preprint Jul 2026

Effects of width-dependent model hyperparameters and $\ell_2$-regularization on the loss landscape of two-layer ReLU networks

Understanding deep neural networks remains a central challenge in machine learning. In particular, the theoretical properties of even two-layer ReLU networks, especially in the presence of weight decay, remain poorly understood. To this end, we derive a sufficient condition on the hyperparameter settings under which the global minima collapse to the zero solution. Interestingly, our experiments reveal that using AdamW as an optimizer prevents the collapse of the learned parameters, whereas using SGD does not, which may help explain the success of AdamW in deep learning training. In addition, when restricting the input dimension to one, we derive an analytical solution for the globally optimal parameter sets of two-layer ReLU networks and show that $\ell_2$-regularization has a width-invariant effect on connectivity, but its dimensionality-reducing effect becomes stronger as the network width increases. These results provide insight into how width-dependent hyperparameters influence the geometry of regularized loss landscapes.

Haruka Eshima, Makoto Yamada · 0 citations
Preprint Aug 2026

Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts

Mixture-of-Experts (MoE) architectures significantly expand model capacity without a proportional increase in computational cost. However, optimizing their hyperparameters---particularly the learning rate---at extreme scales of both model size and token budget via sweeping remains computationally prohibitive. In this paper, we propose a compute-efficient, two-step hyperparameter transfer framework that estimates optimal learning rates for training large MoE models by transferring them across scaling model widths, and subsequently extrapolating to trillion-token horizons. First, we formulate a Maximal Update Parameterization ($\mu$P) adaptation for MoE architectures utilizing Multi-head Latent Attention (MLA) and the Muon optimizer, demonstrating that optimal learning rates transfer consistently across width-scaled models. Second, we extend this transferability along the token dimension by establishing a predictive scaling law. By applying linear regression to the optimal values derived from small proxy models on limited budgets, we successfully extrapolate the ideal learning rate to massive training horizons (e.g., 10 trillion tokens) with high fidelity ($R^2=0.95$). Consequently, this indicates that proxy training on small models is sufficient to determine the optimal learning rate for the extensive training of large-scale MoEs. We apply the proposed methodology to pretrain our foundation model (155B total, 17B active parameters) from scratch, and the stable training and evaluation results validate that optimal configurations for full-scale target models can be accurately predicted with minimal ablation costs.

Nayeon Kim, Hojin Lee, Yunju Bak et al. · 0 citations
Preprint Aug 2026

Accelerated Learning of High Dimensional Functions with a Tensor-Featured Training Network

This work presents a method to accelerate the optimization of learning high dimensional functions using deep neural network (DNN), and studies the effect of adding features which distill pretrained DNN into TNs using a discretize and decompose strategy.

Karl Pierce, Y. Khoo, Haizhao Yang · 0 citations
Preprint Jul 2026

TopoTuner: Topological Finetuning of Large Language Models

TopoTuner is competitive with full fine-tuning while training only 1-2% of the model parameters, and outperforms LoRA in 7 out of 9 model-dataset settings, which can change up to 39.57% of the projection parameters.

Abdulkadir Erol, Yash Mahajan, Vepaul Hariprashad et al. · 0 citations
Review Open access 2026

Transfer Learning Approaches for Small-Scale Datasets

The results confirm that transfer learning is a core solution to the real-world problems that may be limited by available labeled data.

Sivesh Prasana, Karthick eyan · 0 citations
Open access 2026

HIFN-Transformer: Learnable Information-Theoretic Parameters for Interpretable Deep Classification

HIFN-T is presented, a framework extending the Variational Information Bottleneck through four jointly learnable per-layer parameters: information retention, entropy budget, magnitude scaling, and global information gates that generalizes standard VIB as a special case and characterize the role of the entropy budget as an early-training stabilizer that prevents variance explosion before convergence.

Mohammed Tawfik · 0 citations