Maximum Entropy Weighted Independent Set Pooling for Graph Neural Networks

Amirhossein, Nouranizadeh,; Matinkia, Mohammadjavad; Rahmati, Mohammad; Safabakhsh, Reza

doi:10.48550/arxiv.2107.01410

Cited by 3 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we perform the MEWISPool which is a structure-adaptive pooling method to select the crucial sub-structure of brain imaging for every individual. It is incorporated into GNNs in an end-to-end manner and enables effective hierarchical representations of graph 19 . Compared to other graph pooling methods [20][21] , it is designed to preserve the information and connectivity of the graph and reduce redundant information.…”

Section: Maximum Entropy Weighted Independent Set Pooling (Mewispool)mentioning

confidence: 99%

“…The node signals of the graph are the messages to be transmitted and the edges represent the symbols which might be confused at the output due to the presence of noise. This method does not consider the information relationship between nodes and neighbors, but tends to maximize the mutual information between the pooled nodes and the whole set of input nodes 19 .…”

Section: Maximum Entropy Weighted Independent Set Pooling (Mewispool)mentioning

confidence: 99%

See 1 more Smart Citation

Identification of autism spectrum disorder based on MEWISPool and multi-modal learning

Shen¹,

Wang²,

Geng³

2023

Medical Imaging 2023: Computer-Aided Diagnosis

View full text Add to dashboard Cite

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder that impacts how people communicate, interact, behave and learn. Neuroimaging techniques including MRI have been used to not only characterize biomarkers of ASD but also identify individuals with ASD based on analyzing neural structural and functional features, which may assist the precise diagnosis of ASD especially at an early age. Most existing neuroimaging methods for ASD identification focus on a single type of neural measure such as brain functional connections, but ignore the influence of other neural features such as regional activities. There is an increasing need for computational models to use complementary information from multi-modal data for identifying mental disorders. In this work, we propose a framework of graph convolution networks (GCNs) based on maximum entropy weighted independent set pooling (MEWISPool), called MEWISPool-GCN, which not only learns the functional connections and regional activities of the entire brain network, but also integrates non-imaging data such as demographics. Specifically, the graph structure of brain imaging is first downsampled by the MEWISPool method. This structure-adaptive pooling method considers the input graph structure as a noisy communication channel to maximize the mutual information between the input nodes and the pooled nodes. Then, a population graph is constructed in order to further recalibrate the distribution of extracted features using the non-imaging phenotypic information. The feature vectors obtained after pooling are embedded into the nodes of the population graph; and the similarities between non-imaging data are used as edges connecting the nodes. GCN is then employed to learn node embeddings. In the experiment of ASD identification using ABIDE-I dataset, MEWISPool-GCN achieved an accuracy of 87.68% and AUC of 92.89%, which outperformed other related classification methods.

show abstract

Section: Maximum Entropy Weighted Independent Set Pooling (Mewispool)mentioning

confidence: 99%

Section: Maximum Entropy Weighted Independent Set Pooling (Mewispool)mentioning

confidence: 99%

Identification of autism spectrum disorder based on MEWISPool and multi-modal learning

Shen¹,

Wang²,

Geng³

2023

Medical Imaging 2023: Computer-Aided Diagnosis

View full text Add to dashboard Cite

show abstract

“…As shown in table 2, the accuracy is better with the help of global representation. The highest average accuracy 0.865 is achieved by a GNNbased approach with maximum entropy weighted independent set, SetMEWISPool [28] plus global feature. For another graph kernelbased deep learning method, DDGK [1], the best performance is achieved with one-hot embeddings instead of hypergraphlet counting.…”

Section: Enzyme Classificationmentioning

confidence: 99%

Combined topological data analysis and geometric deep learning reveal niches by the quantification of protein binding pockets

Jiang,

Lugo-Martinez

2023

Preprint

View full text Add to dashboard Cite

Protein pockets are essential for many proteins to carry out their functions. Locating and measuring protein pockets as well as studying the anatomy of pockets helps us further understand protein function. Most research studies focus on learning either local or global information from protein structures. However, there is a lack of studies that leverage the power of integrating both local and global representations of these structures. In this work, we combine topological data analysis (TDA) and geometric deep learning (GDL) to analyze the putative protein pockets of enzymes. TDA captures blueprints of the global topological invariant of protein pockets, whereas GDL decomposes the fingerprints to building blocks of these pockets. This integration of local and global views provides a comprehensive and complementary understanding of the protein structural motifs (nichesfor short) within protein pockets. We also analyze the distribution of the building blocks making up the pocket and profile the predictive power of coupling local and global representations for the task of discriminating between enzymes and non-enzymes. We demonstrate that our representation learning framework for macromolecules is particularly useful when the structure is known, and the scenarios heavily rely on local and global information.

show abstract

“…There are several exceptional graph neural networks reported on the PROTEINS and ENZYMES dataset which have gained good scores [16][17][18] . The most prominent ones are hierarchical structure methods employing convolutional layers as well as messagepassing layers 19,20 .…”

Section: Introductionmentioning

confidence: 99%

Introducing CARATE: Finally speaking chemistry through learning hidden wave-function representations on graph-based attention and convolutional neural networks

Kleber

2022

Preprint

View full text Add to dashboard Cite

Computer-Aided Drug Design is advancing to a new era. Recent developments in statistical modelling, including Deep Learning, Machine Learning and high throughput simulations, enable workflows and deductions not achievable 20 years ago. The key interaction for many small molecules is via bio-molecules. The interaction between a small molecule and a biological system therefore manifests itself at multiple time and length scales. While the human chemist quite intuitively grasps the concept of multiple scales, most of the computer technologies do not relate multiple scales easily. Therefore, numerous methods in the realm of computational sciences have been developed. However, up to now it was not clear that the problem of multiple scales is not only a mere matter of computational abilities but even more a matter of accurate representation. Because of the amount of already performed simulations at various scales, deep learning (DL) becomes a viable approach to speed up computations by approximating simulations in a data-driven way. However, to accurately approximate the physical (simulated) properties of a given compound, an accurate, uniform representation is mandatory. Therefore, the biochemical and pharmaceutical encoder (CARATE) is introduced. Furthermore, the regression and classification abilities of CARATE are evaluated against benchmarking datasets (ZINC, ALCHEMY, MCF-7, MOLT-4, YEAST, Enzymes, Proteins) and compared to other baseline approaches.

show abstract

Maximum Entropy Weighted Independent Set Pooling for Graph Neural Networks

Cited by 3 publications

References 35 publications

Identification of autism spectrum disorder based on MEWISPool and multi-modal learning

Identification of autism spectrum disorder based on MEWISPool and multi-modal learning

Combined topological data analysis and geometric deep learning reveal niches by the quantification of protein binding pockets

Introducing CARATE: Finally speaking chemistry through learning hidden wave-function representations on graph-based attention and convolutional neural networks

Contact Info

Product

Resources

About