Abstract: We introduce the variational graph auto-encoder (VGAE), a framework for unsupervised learning on graph-structured data based on the variational auto-encoder (VAE). This model makes use of latent variables and is capable of learning interpretable latent representations for undirected graphs. We demonstrate this model using a graph convolutional network (GCN) encoder and a simple inner product decoder. Our model achieves competitive results on a link prediction task in citation networks. In contrast to most existing models for unsupervised learning on graph-structured data and link prediction, our model can naturally incorporate node features, which significantly improves predictive performance on a number of benchmark datasets.

Abstract: We present a semi-supervised learning framework based on graph embeddings. Given a graph between instances, we train an embedding for each instance to jointly predict the class label and the neighborhood context in the graph. We develop both transductive and inductive variants of our method. In the transductive variant of our method, the class labels are determined by both the learned embeddings and input feature vectors, while in the inductive variant, the embeddings are defined as a parametric function of the feature vectors, so predictions can be made on instances not seen during training. On a large and diverse set of benchmark tasks, including text classification, distantly supervised entity extraction, and entity classification, we show improved performance over many of the existing models.

Abstract: Molecular "fingerprints" encoding structural information are the workhorse of cheminformatics and machine learning in drug discovery applications. However, fingerprint representations necessarily emphasize particular aspects of the molecular structure while ignoring others, rather than allowing the model to make data-driven decisions. We describe molecular "graph convolutions", a machine learning architecture for learning from undirected graphs, specifically small molecules. Graph convolutions use a simple encoding of the molecular graph---atoms, bonds, distances, etc.---which allows the model to take greater advantage of information in the graph structure. Although graph convolutions do not outperform all fingerprint-based methods, they (along with other graph-based methods) represent a new paradigm in ligand-based virtual screening with exciting opportunities for future improvement.

Abstract: In this paper, we propose a novel model for learning graph representations, which generates a low-dimensional vector representation for each vertex by capturing the graph structural information. Different from other previous research efforts, we adopt a random surfing model to capture graph structural information directly, instead of using the sampling-based method for generating linear sequences proposed by Perozzi et al. (2014). The advantages of our approach will be illustrated from both theorical and empirical perspectives. We also give a new perspective for the matrix factorization method proposed by Levy and Goldberg (2014), in which the pointwise mutual information (PMI) matrix is considered as an analytical solution to the objective function of the skip-gram model with negative sampling proposed by Mikolov et al. (2013). Unlike their approach which involves the use of the SVD for finding the low-dimensitonal projections from the PMI matrix, however, the stacked denoising autoencoder is introduced in our model to extract complex features and model non-linearities. To demonstrate the effectiveness of our model, we conduct experiments on clustering and visualization tasks, employing the learned vertex representations as features. Empirical results on datasets of varying sizes show that our model outperforms other stat-of-the-art models in such tasks.

Abstract: Graph-based clustering methods perform clustering on a fixed input data graph. If this initial construction is of low quality then the resulting clustering may also be of low quality. Moreover, existing graph-based clustering methods require post-processing on the data graph to extract the clustering indicators. We address both of these drawbacks by allowing the data graph itself to be adjusted as part of the clustering procedure. In particular, our Constrained Laplacian Rank (CLR) method learns a graph with exactly k connected components (where k is the number of clusters). We develop two versions of this method, based upon the L1-norm and the L2-norm, which yield two new graph-based clustering objectives. We derive optimization algorithms to solve these objectives. Experimental results on synthetic datasets and real-world benchmark datasets exhibit the effectiveness of this new graph-based clustering method.

Abstract: Large networks are becoming a widely used abstraction for studying complex systems in a broad set of disciplines, ranging from social-network analysis to molecular biology and neuroscience. Despite an increasing need to analyze and manipulate large networks, only a limited number of tools are available for this task. Here, we describe the Stanford Network Analysis Platform (SNAP), a general-purpose, high-performance system that provides easy-to-use, high-level operations for analysis and manipulation of large networks. We present SNAP functionality, describe its implementational details, and give performance benchmarks. SNAP has been developed for single big-memory machines, and it balances the trade-off between maximum performance, compact in-memory graph representation, and the ability to handle dynamic graphs in which nodes and edges are being added or removed over time. SNAP can process massive networks with hundreds of millions of nodes and billions of edges. SNAP offers over 140 different graph algorithms that can efficiently manipulate large graphs, calculate structural properties, generate regular and random graphs, and handle attributes and metadata on nodes and edges. Besides being able to handle large graphs, an additional strength of SNAP is that networks and their attributes are fully dynamic; they can be modified during the computation at low cost. SNAP is provided as an open-source library in C++ as well as a module in Python. We also describe the Stanford Large Network Dataset, a set of social and information real-world networks and datasets, which we make publicly available. The collection is a complementary resource to our SNAP software and is widely used for development and benchmarking of graph analytics algorithms.

Abstract: The analysis and optimization of complex systems can be reduced to mathematical problems collectively known as combinatorial optimization. Many such problems can be mapped onto ground-state search problems of the Ising model, and various artificial spin systems are now emerging as promising approaches. However, physical Ising machines have suffered from limited numbers of spin-spin couplings because of implementations based on localized spins, resulting in severe scalability problems. We report a 2000-spin network with all-to-all spin-spin couplings. Using a measurement and feedback scheme, we coupled time-multiplexed degenerate optical parametric oscillators to implement maximum cut problems on arbitrary graph topologies with up to 2000 nodes. Our coherent Ising machine outperformed simulated annealing in terms of accuracy and computation time for a 2000-node complete graph.

Abstract: This paper builds off recent work from Kiperwasser & Goldberg (2016) using neural attention in a simple graph-based dependency parser. We use a larger but more thoroughly regularized parser than other recent BiLSTM-based approaches, with biaffine classifiers to predict arcs and labels. Our parser gets state of the art or near state of the art performance on standard treebanks for six different languages, achieving 95.7% UAS and 94.1% LAS on the most popular English PTB dataset. This makes it the highest-performing graph-based parser on this benchmark---outperforming Kiperwasser Goldberg (2016) by 1.8% and 2.2%---and comparable to the highest performing transition-based parser (Kuncoro et al., 2016), which achieves 95.8% UAS and 94.6% LAS. We also show which hyperparameter choices had a significant effect on parsing accuracy, allowing us to achieve large gains over other graph-based approaches.

Abstract: Molecular “fingerprints” encoding structural information are the workhorse of cheminformatics and machine learning in drug discovery applications. However, fingerprint representations necessarily emphasize particular aspects of the molecular structure while ignoring others, rather than allowing the model to make data-driven decisions. We describe molecular graph convolutions, a machine learning architecture for learning from undirected graphs, specifically small molecules. Graph convolutions use a simple encoding of the molecular graph—atoms, bonds, distances, etc.—which allows the model to take greater advantage of information in the graph structure. Although graph convolutions do not outperform all fingerprint-based methods, they (along with other graph-based methods) represent a new paradigm in ligand-based virtual screening with exciting opportunities for future improvement.

Abstract: This paper builds off recent work from Kiperwasser & Goldberg (2016) using neural attention in a simple graph-based dependency parser. We use a larger but more thoroughly regularized parser than other recent BiLSTM-based approaches, with biaffine classifiers to predict arcs and labels. Our parser gets state of the art or near state of the art performance on standard treebanks for six different languages, achieving 95.7% UAS and 94.1% LAS on the most popular English PTB dataset. This makes it the highest-performing graph-based parser on this benchmark---outperforming Kiperwasser Goldberg (2016) by 1.8% and 2.2%---and comparable to the highest performing transition-based parser (Kuncoro et al., 2016), which achieves 95.8% UAS and 94.6% LAS. We also show which hyperparameter choices had a significant effect on parsing accuracy, allowing us to achieve large gains over other graph-based approaches.

Abstract: Graph-based approaches have been successful in unsupervised and semi-supervised learning. In this paper, we focus on the real-world applications where the same instance can be represented by multiple heterogeneous features. The key point of utilizing the graph-based knowledge to deal with this kind of data is to reasonably integrate the different representations and obtain the most consistent manifold with the real data distributions. In this paper, we propose a novel framework via the reformulation of the standard spectral learning model, which can be used for multiview clustering and semisupervised tasks. Unlike other methods in the literature, the proposed methods can learn an optimal weight for each graph automatically without introducing an additive parameter as previous methods do. Furthermore, our objective under semisupervised learning is convex and the global optimal result will be obtained. Extensive empirical results on different real-world data sets demonstrate that the proposed methods achieve comparable performance with the state-of-the-art approaches and can be used more practically.

Abstract: Low-rank representation (LRR) has recently attracted a great deal of attention due to its pleasing efficacy in exploring low-dimensional subspace structures embedded in data. For a given set of observed data corrupted with sparse errors, LRR aims at learning a lowest-rank representation of all data jointly. LRR has broad applications in pattern recognition, computer vision and signal processing. In the real world, data often reside on low-dimensional manifolds embedded in a high-dimensional ambient space. However, the LRR method does not take into account the non-linear geometric structures within data, thus the locality and similarity information among data may be missing in the learning process. To improve LRR in this regard, we propose a general Laplacian regularized low-rank representation framework for data representation where a hypergraph Laplacian regularizer can be readily introduced into, i.e., a Non-negative Sparse Hyper-Laplacian regularized LRR model (NSHLRR). By taking advantage of the graph regularizer, our proposed method not only can represent the global low-dimensional structures, but also capture the intrinsic non-linear geometric information in data. The extensive experimental results on image clustering, semi-supervised image classification and dimensionality reduction tasks demonstrate the effectiveness of the proposed method.

Abstract: By taking the semantic object parsing task as an exemplar application scenario, we propose the Graph Long Short-Term Memory (Graph LSTM) network, which is the generalization of LSTM from sequential data or multi-dimensional data to general graph-structured data. Particularly, instead of evenly and fixedly dividing an image to pixels or patches in existing multi-dimensional LSTM structures (e.g., Row, Grid and Diagonal LSTMs), we take each arbitrary-shaped superpixel as a semantically consistent node, and adaptively construct an undirected graph for each image, where the spatial relations of the superpixels are naturally used as edges. Constructed on such an adaptive graph topology, the Graph LSTM is more naturally aligned with the visual patterns in the image (e.g., object boundaries or appearance similarities) and provides a more economical information propagation route. Furthermore, for each optimization step over Graph LSTM, we propose to use a confidence-driven scheme to update the hidden and memory states of nodes progressively till all nodes are updated. In addition, for each node, the forgets gates are adaptively learned to capture different degrees of semantic correlation with neighboring nodes. Comprehensive evaluations on four diverse semantic object parsing datasets well demonstrate the significant superiority of our Graph LSTM over other state-of-the-art solutions.

Abstract: We propose a framework that learns the graph structure underlying a set of smooth signals. Given a real m by n matrix X whose rows reside on the vertices of an unknown graph, we learn the edge weights w under the smoothness assumption that trace(X^TLX) is small. We show that the problem is a weighted l-1 minimization that leads to naturally sparse solutions. We point out how known graph learning or construction techniques fall within our framework and propose a new model that performs better than the state of the art in many settings. We present efficient, scalable primal-dual based algorithms for both our model and the previous state of the art, and evaluate their performance on artificial and real data.

Abstract: We study the problem of visualizing large-scale and high-dimensional data in a low-dimensional (typically 2D or 3D) space. Much success has been reported recently by techniques that first compute a similarity structure of the data points and then project them into a low-dimensional space with the structure preserved. These two steps suffer from considerable computational costs, preventing the state-of-the-art methods such as the t-SNE from scaling to large-scale and high-dimensional data (e.g., millions of data points and hundreds of dimensions). We propose the LargeVis, a technique that first constructs an accurately approximated K-nearest neighbor graph from the data and then layouts the graph in the low-dimensional space. Comparing to t-SNE, LargeVis significantly reduces the computational cost of the graph construction step and employs a principled probabilistic model for the visualization step, the objective of which can be effectively optimized through asynchronous stochastic gradient descent with a linear time complexity. The whole procedure thus easily scales to millions of high-dimensional data points. Experimental results on real-world data sets demonstrate that the LargeVis outperforms the state-of-the-art methods in both efficiency and effectiveness. The hyper-parameters of LargeVis are also much more stable over different data sets.

Abstract: Given a bipartite graph of users and the products that they review, or followers and followees, how can we detect fake reviews or follows? Existing fraud detection methods (spectral, etc.) try to identify dense subgraphs of nodes that are sparsely connected to the remaining graph. Fraudsters can evade these methods using camouflage, by adding reviews or follows with honest targets so that they look "normal". Even worse, some fraudsters use hijacked accounts from honest users, and then the camouflage is indeed organic. Our focus is to spot fraudsters in the presence of camouflage or hijacked accounts. We propose FRAUDAR, an algorithm that (a) is camouflage-resistant, (b) provides upper bounds on the effectiveness of fraudsters, and (c) is effective in real-world data. Experimental results under various attacks show that FRAUDAR outperforms the top competitor in accuracy of detecting both camouflaged and non-camouflaged fraud. Additionally, in real-world experiments with a Twitter follower-followee graph of 1.47 billion edges, FRAUDAR successfully detected a subgraph of more than 4000 detected accounts, of which a majority had tweets showing that they used follower-buying services.

Abstract: Large networks are becoming a widely used abstraction for studying complex systems in a broad set of disciplines, ranging from social network analysis to molecular biology and neuroscience. Despite an increasing need to analyze and manipulate large networks, only a limited number of tools are available for this task. Here, we describe Stanford Network Analysis Platform (SNAP), a general-purpose, high-performance system that provides easy to use, high-level operations for analysis and manipulation of large networks. We present SNAP functionality, describe its implementational details, and give performance benchmarks. SNAP has been developed for single big-memory machines and it balances the trade-off between maximum performance, compact in-memory graph representation, and the ability to handle dynamic graphs where nodes and edges are being added or removed over time. SNAP can process massive networks with hundreds of millions of nodes and billions of edges. SNAP offers over 140 different graph algorithms that can efficiently manipulate large graphs, calculate structural properties, generate regular and random graphs, and handle attributes and meta-data on nodes and edges. Besides being able to handle large graphs, an additional strength of SNAP is that networks and their attributes are fully dynamic, they can be modified during the computation at low cost. SNAP is provided as an open source library in C++ as well as a module in Python. We also describe the Stanford Large Network Dataset, a set of social and information real-world networks and datasets, which we make publicly available. The collection is a complementary resource to our SNAP software and is widely used for development and benchmarking of graph analytics algorithms.

Abstract: We present Reactive Vega, a system architecture that provides the first robust and comprehensive treatment of declarative visual and interaction design for data visualization. Starting from a single declarative specification, Reactive Vega constructs a dataflow graph in which input data, scene graph elements, and interaction events are all treated as first-class streaming data sources. To support expressive interactive visualizations that may involve time-varying scalar, relational, or hierarchical data, Reactive Vega's dataflow graph can dynamically re-write itself at runtime by extending or pruning branches in a data-driven fashion. We discuss both compile- and run-time optimizations applied within Reactive Vega, and share the results of benchmark studies that indicate superior interactive performance to both D3 and the original, non-reactive Vega system.

Abstract: In this work, we propose a novel approach to video segmentation that operates in bilateral space. We design a new energy on the vertices of a regularly sampled spatiotemporal bilateral grid, which can be solved efficiently using a standard graph cut label assignment. Using a bilateral formulation, the energy that we minimize implicitly approximates long-range, spatio-temporal connections between pixels while still containing only a small number of variables and only local graph edges. We compare to a number of recent methods, and show that our approach achieves state-of-the-art results on multiple benchmarks in a fraction of the runtime. Furthermore, our method scales linearly with image size, allowing for interactive feedback on real-world high resolution video.

Abstract: With the large volume of new information created every day, determining the validity of information in a knowledge graph and filling in its missing parts are crucial tasks for many researchers and practitioners. To address this challenge, a number of knowledge graph completion methods have been developed using low-dimensional graph embeddings. Although researchers continue to improve these models using an increasingly complex feature space, we show that simple changes in the architecture of the underlying model can outperform state-of-the-art models without the need for complex feature engineering. In this work, we present a shared variable neural network model called ProjE that fills-in missing information in a knowledge graph by learning joint embeddings of the knowledge graph's entities and edges, and through subtle, but important, changes to the standard loss function. In doing so, ProjE has a parameter size that is smaller than 11 out of 15 existing methods while performing $37\%$ better than the current-best method on standard datasets. We also show, via a new fact checking task, that ProjE is capable of accurately determining the veracity of many declarative statements.

Abstract: We study the problem of visualizing large-scale and high-dimensional data in a low-dimensional (typically 2D or 3D) space. Much success has been reported recently by techniques that first compute a similarity structure of the data points and then project them into a low-dimensional space with the structure preserved. These two steps suffer from considerable computational costs, preventing the state-of-the-art methods such as the t-SNE from scaling to large-scale and high-dimensional data (e.g., millions of data points and hundreds of dimensions). We propose the LargeVis, a technique that first constructs an accurately approximated K-nearest neighbor graph from the data and then layouts the graph in the low-dimensional space. Comparing to t-SNE, LargeVis significantly reduces the computational cost of the graph construction step and employs a principled probabilistic model for the visualization step, the objective of which can be effectively optimized through asynchronous stochastic gradient descent with a linear time complexity. The whole procedure thus easily scales to millions of high-dimensional data points. Experimental results on real-world data sets demonstrate that the LargeVis outperforms the state-of-the-art methods in both efficiency and effectiveness. The hyper-parameters of LargeVis are also much more stable over different data sets.

Abstract: Graph matching (GM) is a fundamental problem in computer science, and it plays a central role to solve correspondence problems in computer vision. GM problems that incorporate pairwise constraints can be formulated as a quadratic assignment problem (QAP). Although widely used, solving the correspondence problem through GM has two main limitations: (1) the QAP is NP-hard and difficult to approximate; (2) GM algorithms do not incorporate geometric constraints between nodes that are natural in computer vision problems. To address aforementioned problems, this paper proposes factorized graph matching (FGM). FGM factorizes the large pairwise affinity matrix into smaller matrices that encode the local structure of each graph and the pairwise affinity between edges. Four are the benefits that follow from this factorization: (1) There is no need to compute the costly (in space and time) pairwise affinity matrix; (2) The factorization allows the use of a path-following optimization algorithm, that leads to improved optimization strategies and matching performance; (3) Given the factorization, it becomes straight-forward to incorporate geometric transformations (rigid and non-rigid) to the GM problem. (4) Using a matrix formulation for the GM problem and the factorization, it is easy to reveal commonalities and differences between different GM methods. The factorization also provides a clean connection with other matching algorithms such as iterative closest point; Experimental results on synthetic and real databases illustrate how FGM outperforms state-of-the-art algorithms for GM. The code is available at http://humansensing.cs.cmu.edu/fgm .

Abstract: Many graph-based semi-supervised learning methods for large datasets have been proposed to cope with the rapidly increasing size of data, such as Anchor Graph Regularization (AGR). This model builds a regularization framework by exploring the underlying structure of the whole dataset with both datapoints and anchors. Nevertheless, AGR still has limitations in its two components: (1) in anchor graph construction, the estimation of the local weights between each datapoint and its neighboring anchors could be biased and relatively slow; and (2) in anchor graph regularization, the adjacency matrix that estimates the relationship between datapoints, is not sufficiently effective. In this paper, we develop an Efficient Anchor Graph Regularization (EAGR) by tackling these issues. First, we propose a fast local anchor embedding method, which reformulates the optimization of local weights and obtains an analytical solution. We show that this method better reconstructs datapoints with anchors and speeds up the optimizing process. Second, we propose a new adjacency matrix among anchors by considering the commonly linked datapoints, which leads to a more effective normalized graph Laplacian over anchors. We show that, with the novel local weight estimation and normalized graph Laplacian, EAGR is able to achieve better classification accuracy with much less computational costs. Experimental results on several publicly available datasets demonstrate the effectiveness of our approach.

Abstract: Dynamic environments have obstacles that unpredictably appear, disappear, or move. We present the first sampling-based replanning algorithm that is asymptotically optimal and single-query (designed for situation in which a priori offline computation is unavailable). Our algorithm, RRTX, refines and repairs the same search-graph over the entire duration of navigation (in contrast to previous single-query replanning algorithms that prune and then regrow some or all of the search-tree). Whenever obstacles change and/or the robot moves, a graph rewiring cascade quickly remodels the existing search-graph and repairs its shortest-path-to-goal sub-tree to reflect the new information. Both graph and tree are built directly in the robot’s state-space; thus, the resulting plan(s) respect the kinematics of the robot and continue to improve during navigation. RRTX is probabilistically complete and makes no distinction between local and global planning, yet it reacts quickly enough for real-time high-speed navigation ...

Abstract: We describe a method to automatically detect contours, i.e. lines along which the surface orientation sharply changes, in large-scale outdoor point clouds. Contours are important intermediate features for structuring point clouds and converting them into high-quality surface or solid models, and are extensively used in graphics and mapping applications. Yet, detecting them in unstructured, inhomogeneous point clouds turns out to be surprisingly difficult, and existing line detection algorithms largely fail. We approach contour extraction as a two-stage discriminative learning problem. In the first stage, a contour score for each individual point is predicted with a binary classifier, using a set of features extracted from the point's neighborhood. The contour scores serve as a basis to construct an overcomplete graph of candidate contours. The second stage selects an optimal set of contours from the candidates. This amounts to a further binary classification in a higher-order MRF, whose cliques encode a preference for connected contours and penalize loose ends. The method can handle point clouds > 107 points in a couple of minutes, and vastly outperforms a baseline that performs Canny-style edge detection on a range image representation of the point cloud.

Abstract: Summary Community detection in networks is a key exploratory tool with applications in a diverse set of areas, ranging from finding communities in social and biological networks to identifying link farms in the World Wide Web. The problem of finding communities or clusters in a network has received much attention from statistics, physics and computer science. However, most clustering algorithms assume knowledge of the number of clusters k. We propose to determine k automatically in a graph generated from a stochastic block model by using a hypothesis test of independent interest. Our main contribution is twofold; first, we theoretically establish the limiting distribution of the principal eigenvalue of the suitably centred and scaled adjacency matrix and use that distribution for our test of the hypothesis that a random graph is of Erdős–Renyi (noise) type. Secondly, we use this test to design a recursive bipartitioning algorithm, which naturally uncovers nested community structure. Using simulations and quantifiable classification tasks on real world networks with ground truth, we show that our algorithm outperforms state of the art methods.

Abstract: Identification of module or community structures is important for characterizing and understanding complex systems. While designed with different objectives, i.e., stochastic models for regeneration and modularity maximization models for discrimination, both these two types of model look for low-rank embedding to best represent and reconstruct network topology. However, the mapping through such embedding is linear, whereas real networks have various nonlinear features, making these models less effective in practice. Inspired by the strong representation power of deep neural networks, we propose a novel nonlinear reconstruction method by adopting deep neural networks for representation. We then extend the method to a semi-supervised community detection algorithm by incorporating pairwise constraints among graph nodes. Extensive experimental results on synthetic and real networks show that the new methods are effective, outperforming most state-of-the-art methods for community detection.

Abstract: In this paper, we propose a paradigm shift in representing and optimizing logic by using only majority (MAJ) and inversion (INV) functions as basic operations. We represent logic functions by majority-inverter graph (MIG): a directed acyclic graph consisting of three-input majority nodes and regular/complemented edges. We optimize MIGs via a new Boolean algebra, based exclusively on majority and inversion operations, that we formally axiomatize in this paper. As a complement to MIG algebraic optimization, we develop powerful Boolean methods exploiting global properties of MIGs, such as bit-error masking. MIG algebraic and Boolean methods together attain very high optimization quality. Considering the set of IWLS’05 benchmarks, our MIG optimizer (MIGhty) enables a 7% depth reduction in LUT-6 circuits mapped by ABC while also reducing size and power activity, with respect to similar and-inverter graph (AIG) optimization. Focusing on arithmetic intensive benchmarks instead, MIGhty enables a 16% depth reduction in LUT-6 circuits mapped by ABC, again with respect to similar AIG optimization. Employed as front-end to a delay-critical 22-nm application-specified integrated circuit flow (logic synthesis + physical design) MIGhty reduces the average delay/area/power by 13%/4%/3%, respectively, over 31 academic and industrial benchmarks. We also demonstrate delay/area/power improvements by 10%/10%/5% for a commercial FPGA flow.