Generalization error of graph neural networks in the mean-field regime

Generalization error of graph neural networks in the mean-field regime

This work provides a theoretical framework for assessing the generalization error of graph neural networks in the over-parameterized regime, where the number of parameters surpasses the quantity of data points. We explore two widely utilized types of graph neural networks: graph convolutional neural...

Volledige beschrijving

Bibliografische gegevens
Hoofdauteurs:	Aminian, G, He, Y, Reinert, G, Szpruch, L, Cohen, S
Formaat:	Conference item
Taal:	English
Gepubliceerd in:	Proceedings of Machine Learning Research 2024

Gelijkaardige items

MSGNN: a spectral graph neural network based on a novel magnetic signed Laplacian
door: He, Y, et al.
Gepubliceerd in: (2022)

GNNRank: learning global rankings from pairwise comparisons via directed graph neural networks
door: He, Y, et al.
Gepubliceerd in: (2022)

Graph neural networks for network analysis
door: He, Y
Gepubliceerd in: (2024)

GDRMA: Graph Neural Networks for Document Retrievals With Mean Aggregation
door: Shigeru Maya
Gepubliceerd in: (2024-01-01)

Physics-Informed Graph Neural Operator for Mean Field Games on Graph: A Scalable Learning Approach
door: Xu Chen, et al.
Gepubliceerd in: (2024-03-01)

Mutual information of neural network initialisations: mean field approximations
door: Tanner, J, et al.
Gepubliceerd in: (2021)

Computational and topological properties of neural networks by means of graph-theoretic parameters
door: Asad Khan, et al.
Gepubliceerd in: (2023-03-01)

Proximal mean-field for neural network quantization
door: Ajanthan, T, et al.
Gepubliceerd in: (2020)

Mean field analysis of deep neural networks
door: Sirignano, J, et al.
Gepubliceerd in: (2021)

A limit order book model for latency arbitrage
door: Cohen, SN, et al.
Gepubliceerd in: (2012)

A limit order book model for latency arbitrage
door: Cohen, SN, et al.
Gepubliceerd in: (2011)

Asymmetric mean-field neural networks for multiprocessor scheduling /
door: 286373 Hellstrom, Benjamin J., et al.

Graph coloring with physics-inspired graph neural networks
door: Martin J. A. Schuetz, et al.
Gepubliceerd in: (2022-11-01)

Error Analysis of Terrestrial Laser Scanning Data by Means of Spherical Statistics and 3D Graphs
door: Pedro Arias, et al.
Gepubliceerd in: (2010-11-01)

Graph neural network with knowledge graph
door: Ang, Qi Xuan
Gepubliceerd in: (2020)

Improved Modeling and Generalization Capabilities of Graph Neural Networks With Legendre Polynomials
door: Jiali Chen, et al.
Gepubliceerd in: (2023-01-01)

Identifiability in inverse reinforcement learning
door: Cao, H, et al.
Gepubliceerd in: (2021)

Graph neural networks
door: Lian, Ran
Gepubliceerd in: (2023)

AgraSSt: approximate graph Stein statistics for interpretable assessment of implicit graph generators
door: Xu, W, et al.
Gepubliceerd in: (2023)

On Neural Networks Fitting, Compression, and Generalization Behavior via Information-Bottleneck-like Approaches
door: Zhaoyan Lyu, et al.
Gepubliceerd in: (2023-07-01)

Complement of the generalized total graph of fields
door: T. Tamizh Chelvam, et al.
Gepubliceerd in: (2020-09-01)

A mean field theory learning algorithm for neural networks /
door: 323731 Peterson, Carsten, et al.

Graph Information Vanishing Phenomenon in Implicit Graph Neural Networks
door: Silu He, et al.
Gepubliceerd in: (2024-08-01)

Mean-fields and neural masses.
door: Karl Friston
Gepubliceerd in: (2008-08-01)

Node co-activations as a means of error detection—Towards fault-tolerant neural networks
door: Lalli Myllyaho, et al.
Gepubliceerd in: (2022-09-01)

Convolutional Neural Network Outperforms Graph Neural Network on the Spatially Variant Graph Data
door: Anna Boronina, et al.
Gepubliceerd in: (2023-05-01)

Fractional integral inequalities and error estimates of generalized mean differences
door: Muhammad Samraiz, et al.
Gepubliceerd in: (2024-05-01)

Black-box model risk in finance
door: Cohen, SN, et al.
Gepubliceerd in: (2023)

Benchmarking graph neural networks for materials chemistry
door: Victor Fung, et al.
Gepubliceerd in: (2021-06-01)

Graph neural network coarse-grain force field for the molecular crystal RDX
door: Brian H. Lee, et al.
Gepubliceerd in: (2024-09-01)

Predicting stress, strain and deformation fields in materials and structures with graph neural networks
door: Marco Maurizi, et al.
Gepubliceerd in: (2022-12-01)

Mean field analysis of neural networks: a law of large numbers
door: Sirignano, J, et al.
Gepubliceerd in: (2020)

Mean field analysis of neural networks: a central limit theorem
door: Sirignano, J, et al.
Gepubliceerd in: (2019)

Federated graph neural network
door: Koh, Tat You @ Arthur
Gepubliceerd in: (2021)

Survey on Graph Neural Networks
door: Georgios Gkarmpounis, et al.
Gepubliceerd in: (2024-01-01)

Schatten Graph Neural Networks
door: Youfa Liu, et al.
Gepubliceerd in: (2022-01-01)

Graph neural networks with a distribution of parametrized graphs
door: Lee, See Hian, et al.
Gepubliceerd in: (2024)

Role of stochastic noise and generalization error in the time propagation of neural-network quantum states
door: Damian Hofmann, Giammarco Fabiani, Johan H. Mentink, Giuseppe Carleo, Michael A. Sentef
Gepubliceerd in: (2022-05-01)

Optimizing Graph Neural Network Training on Large Graphs
door: Stathas, Nickolas
Gepubliceerd in: (2022)

Circuit2Graph: Circuits With Graph Neural Networks
door: Yusuke Yamakaji, et al.
Gepubliceerd in: (2024-01-01)