Power Iterated Color Refinement

Kersting, Kristian; Mladenov, Martin; Garnett, Roman; Grohe, Martin

doi:10.1609/aaai.v28i1.8992

Cited by 13 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In summary, our graph convolution model (1) effectively mimics the behavior of two popular graph kernels, which helps explain its graph-level classification performance. The connection between discrete WL, continuous optimization, and random walks have also been studied in (Boldi et al 2006;Kersting et al 2014;Kipf and Welling 2016) etc.…”

Section: Connection With Weisfeiler-lehman Subtree Kernelmentioning

confidence: 99%

An End-to-End Deep Learning Architecture for Graph Classification

Zhang

Cui

Neumann

et al. 2018

AAAI

873

215

View full text Add to dashboard Cite

Neural networks are typically designed to deal with data in tensor forms. In this paper, we propose a novel neural network architecture accepting graphs of arbitrary structure. Given a dataset containing graphs in the form of (G,y) where G is a graph and y is its class, we aim to develop neural networks that read the graphs directly and learn a classification function. There are two main challenges: 1) how to extract useful features characterizing the rich information encoded in a graph for classification purpose, and 2) how to sequentially read a graph in a meaningful and consistent order. To address the first challenge, we design a localized graph convolution model and show its connection with two graph kernels. To address the second challenge, we design a novel SortPooling layer which sorts graph vertices in a consistent order so that traditional neural networks can be trained on the graphs. Experiments on benchmark graph classification datasets demonstrate that the proposed architecture achieves highly competitive performance with state-of-the-art graph kernels and other graph neural network methods. Moreover, the architecture allows end-to-end gradient-based training with original graphs, without the need to first transform graphs into vectors.

show abstract

Section: Connection With Weisfeiler-lehman Subtree Kernelmentioning

confidence: 99%

An End-to-End Deep Learning Architecture for Graph Classification

Zhang

Cui

Neumann

et al. 2018

AAAI

873

215

View full text Add to dashboard Cite

show abstract

“…In addition, as WL requires reading and sorting of the vertices' signature strings, it becomes computationally expensive since the signature strings can be very long for nodes with high degrees. Fast hashing-based WL algorithms were proposed [25,37] which map unique signature strings to unique real values. To deal with issues mentioned above, we borrowed the Pallete-WL algorithm [25] in which it can take advantage of vertex ordering capabi-lity of WL while capturing the core information of each sub-graph (i.e.…”

Section: Sub-graph Pattern Encodingmentioning

confidence: 99%

Drug-target interaction prediction using semi-bipartite graph model and deep learning

Manoochehri

Nourani

2020

BMC Bioinformatics

View full text Add to dashboard Cite

Background Identifying drug-target interaction is a key element in drug discovery. In silico prediction of drug-target interaction can speed up the process of identifying unknown interactions between drugs and target proteins. In recent studies, handcrafted features, similarity metrics and machine learning methods have been proposed for predicting drug-target interactions. However, these methods cannot fully learn the underlying relations between drugs and targets. In this paper, we propose anew framework for drug-target interaction prediction that learns latent features from drug-target interaction network. Results We present a framework to utilize the network topology and identify interacting and non-interacting drug-target pairs. We model the problem as a semi-bipartite graph in which we are able to use drug-drug and protein-protein similarity in a drug-protein network. We have then used a graph labeling method for vertex ordering in our graph embedding process. Finally, we employed deep neural network to learn the complex pattern of interacting pairs from embedded graphs. We show our approach is able to learn sophisticated drug-target topological features and outperforms other state-of-the-art approaches. Conclusions The proposed learning model on semi-bipartite graph model, can integrate drug-drug and protein-protein similarities which are semantically different than drug-protein information in a drug-target interaction network. We show our model can determine interaction likelihood for each drug-target pair and outperform other heuristics.

show abstract

“…Second, the low-rank Boolean matrix factorization used for relational approximations can be applied to any graph structure, including graphical models. Third, color passing techniques for exact symmetries operate on propositional models (Kersting, Ahmadi, and Natarajan 2009;Kersting et al 2014). Combined with early stopping, they can output approximate variable orbits.…”

Section: Lifted Metropolis-hastingsmentioning

confidence: 99%

“…There exist several techniques to compute the exact symmetries of a graphical model and construct G; see (Niepert 2012b;Bui, Huynh, and Riedel 2013). The color refinement algorithm is also well-studied in lifted inference (Kersting et al 2014). It can find (exact) orbits of random variables for a slightly weaker notion of symmetry, called fractional automorphism.…”

Section: Approximate Symmetriesmentioning

confidence: 99%

“…The approach combines a base MCMC algorithm such as the Gibbs sampler with the Metropolis chain that performs jumps in the symmetric model. The novel framework allows us to utilize work on approximate symmetries such as color passing algorithms (Kersting et al 2014) and lowrank Boolean matrix approximations (Van den Broeck and Darwiche 2013) while producing unbiased probability estimates. We identify properties of an approximate symmetry group that make it suitable for the novel lifted Metropolis-Hastings approach.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Lifted Probabilistic Inference for Asymmetric Graphical Models

Broeck

Niepert

2015

AAAI

View full text Add to dashboard Cite

Lifted probabilistic inference algorithms have been successfully applied to a large number of symmetric graphical models. Unfortunately, the majority of real-world graphical models is asymmetric. This is even the case for relational representations when evidence is given. Therefore, more recent work in the community moved to making the models symmetric and then applying existing lifted inference algorithms. However, this approach has two shortcomings. First, all existing over-symmetric approximations require a relational representation such as Markov logic networks. Second, the induced symmetries often change the distribution significantly, making the computed probabilities highly biased. We present a framework for probabilistic sampling-based inference that only uses the induced approximate symmetries to propose steps in a Metropolis-Hastings style Markov chain. The framework, therefore, leads to improved probability estimates while remaining unbiased. Experiments demonstrate that the approach outperforms existing MCMC algorithms.

show abstract

Power Iterated Color Refinement

Cited by 13 publications

References 20 publications

An End-to-End Deep Learning Architecture for Graph Classification

An End-to-End Deep Learning Architecture for Graph Classification

Drug-target interaction prediction using semi-bipartite graph model and deep learning

Lifted Probabilistic Inference for Asymmetric Graphical Models

Contact Info

Product

Resources

About