TF-GNN: Graph Neural Networks in TensorFlow

Ferludin, Oleksandr; Eigenwillig, Arno; Blais, Martin; Zelle, Dustin; Pfeifer, Jan; Sánchez‐González, Álvaro; Li, Sibon; Abu-El-Haija, Sami; Battaglia, Peter W.; Bulut, Neslihan; Halcrow, Jonathan; Almeida, Filipe Miguel Gonçalves de; Lattanzi, Silvio; Linhares, André; Brandon, Mayer,; Mirrokni, Vahab; Palowitch, John; Paradkar, Mihir; She, J.; Tsitsulin, Anton; Villela, Kevin; Wang, Lisa; Wong, David T.; Perozzi, Bryan

doi:10.48550/arxiv.2207.03522

Cited by 3 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 shows detailed configurations of the GAT-LSTM and ANN-LSTM models used in this experiment. The neural network model is implemented on the TensorFlow platform [34], and graph learning is realized using the TensorFlow Graph Neural Networks library [35]. All models are trained on a single NVIDIA GeForce RTX 3090 GPU with 24 GB memory.…”

Section: Case Studiesmentioning

confidence: 99%

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Ma,

Zhao,

Zhou

et al. 2024

Mathematics

View full text Add to dashboard Cite

Inverse modeling can estimate uncertain parameters in subsurface reservoirs and provide reliable numerical models for reservoir development and management. The traditional simulation-based inversion method usually requires numerous numerical simulations, which is time-consuming. Recently, deep learning-based surrogate models have been widely studied as an alternative to numerical simulation, which can significantly improve the solving efficiency of inversion. However, for reservoirs with complex fracture distribution, constructing the surrogate model of numerical simulation presents a significant challenge. In this work, we present a deep graph learning-based surrogate model for inverse modeling of fractured reservoirs. Specifically, the proposed surrogate model integrates the graph attention mechanisms to extract features of fracture network in reservoirs. The graph learning can retain the discrete characteristics and structural information of the fracture network. The extracted features are subsequently integrated with a multi-layer recurrent neural network model to predict the production dynamics of wells. A surrogate-based inverse modeling workflow is then developed by combining the surrogate model with the differential evolutionary algorithm. Numerical studies performed on a synthetic naturally fractured reservoir model with multi-scale fractures illustrate the performance of the proposed methods. The results demonstrate that the proposed surrogate model exhibits promising generalization performance of production prediction. Compared with tens of thousands of numerical simulations required by the simulation-based inverse modeling method, the proposed surrogate-based method only requires 1000 to 1500 numerical simulations, and the solution efficiency can be improved by ten times.

show abstract

Section: Case Studiesmentioning

confidence: 99%

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Ma,

Zhao,

Zhou

et al. 2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…In addition, the irregular nature and instability of graph data pose several computational challenges [20]. Current GNN implementations for learning these complex datasets use frameworks such as the Deep Graph Library (DGL) [16], PyTorch Geometric (PyG) [21], and TensorFlow GNN [22]. These frameworks provide software support for graph neural networks on CPUs and GPUs, which are often used for Convolutional Neural Networks (CNNs) and other well-known methods [23,24].…”

Section: Introductionmentioning

confidence: 99%

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

Kose,

Nunez-Yanez,

Piechocki

et al. 2024

Preprint

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) are powerful models capable of managing intricate connections in non-Euclidean data, such as social networks, physical systems, chemical structures, and communication networks. Despite their effectiveness, the large-scale and complex nature of graph data demands substantial computational resources and high performance during both training and inference stages, presenting significant challenges, particularly in the context of embedded systems. Recent studies on GNNs have investigated both software and hardware solutions to enhance computational efficiency. Earlier studies on Deep Neural Networks (DNNs) have indicated that methods like reconfigurable hardware and quantization are beneficial in addressing these issues. Unlike DNN research, studies on efficient computational methods for GNNs are less developed and require more exploration. This survey reviews the latest developments in quantization and FPGA-based acceleration for GNNs, showcasing the capabilities of reconfigurable systems (often FPGAs) to offer customized solutions in environments marked by significant sparsity and the necessity for dynamic load management. It also emphasizes the role of quantization in reducing both computational and memory demands through the use of fixed-point arithmetic and streamlined vector formats. This paper concentrates on low-power, resource-limited devices over general hardware accelerators and reviews research applicable to embedded systems. Additionally, it provides a detailed discussion of potential research gaps, foundational knowledge, obstacles, and prospective future directions.

show abstract