Extends graph factor system to quasi-median graphs.
problem Constraint relaxation for combinatorial HHS machinery.
method Relaxing domain constraints on combinatorial HHS machinery and extending factor system to quasi-median graphs.
result Factor system applied to quasi-median graphs.
The paper analyzes structured prediction with new margin guarantees and learning algorithms.
problem Structured prediction with arbitrary factor graphs and complex loss functions.
method Data-dependent margin guarantees and Voted Risk Minimization principle.
result New learning bounds and algorithms (VCRF, StructBoost) for complex factor graphs.
Complex tensor factorization improves knowledge graph completion.
problem Automatically understanding and predicting missing relationships in large knowledge graphs.
method Use of complex-valued embeddings and unitary diagonalization.
result Complex embeddings lead to scalable and expressive models that outperform existing methods.
New complex connects graph separability to group properties.
problem Understanding separability of graph fundamental groups.
method Introducing separability complex and proving its properties.
result Separability complex has infinite diameter and is nonhyperbolic.
Tensor variable elimination for plated factor graphs enables exact inference in models with repeated structure.
problem Efficient inference in models with repeated structure.
method Generalized variable elimination to tensor variable elimination on plated factor graphs.
result Tractable inference for a class of plated factor graphs.
We study computational and sample complexity of parameter and structure learning in graphical models. Our main result shows that the class of factor graphs with bounded factor size and bounded connectivity can be learned in polynomial time and polynomial number of samples, assuming that the data is generated by a netwo…
This work provides minimax bounds for structured prediction models.
problem Limited understanding of necessary sample complexity for structured prediction.
method Analysis of factor-graph inference models for structured prediction.
result Characterization of necessary sample complexity for any algorithm.
Improved prediction accuracy in matrix factorization using graph-based priors.
problem Graph side-information may not align with latent-feature relations in matrix completion.
method Identify and remove 'contested' edges using graphical lasso approximation, maintaining linear scalability.
result Improved prediction accuracy with fewer graph edges, demonstrating the often inaccurate nature of graph side-information.
The paper defines complexes for RAAGs connecting buildings and free factor complexes, proving their homotopy Cohen-Macaulay properties.
problem Defining and analyzing complexes for RAAGs to understand their structure.
method Defining simplicial complexes from RAAG outer automorphism groups, using coset complexes and decompositions.
result These complexes are homotopy Cohen-Macaulay and homotopy equivalent to spheres.
EPFGNN models graph connections for better node classification.
problem Graph node classification issues due to feature aggregation.
method EPFGNN models graph as a Markov Random Field with explicit pairwise factors and a GNN backbone.
result EPFGNN improves semi-supervised node classification performance.
Novel CGTF model for recommender systems and community detection from coupled graphs and tensors.
problem Lack of effective methods for analyzing multiple information repositories with graph side information.
method Coupled Graph-Tensor Factorization (CGTF) with ADMM for nonnegative factor recovery.
result CGTF model successfully detects communities even with missing graph links.
Method detects communities in networks using matrix factorization.
problem Community detection in complex networks.
method Orthogonal symmetric non-negative matrix tri-factorization of the normalized Laplacian matrix.
result Consistent for community detection in graphs from stochastic block models.
H-GAT improves stock selection by capturing complex higher-order stock relations and integrating both technical and fundamental analysis.
problem Stock selection difficulty and lack of comprehensive analysis.
method Higher-order Graph Attention Network (H-GAT) that incorporates both technical and fundamental analysis.
result H-GAT outperforms existing methods in stock selection metrics.
Learned factor graphs improve inference from time sequences using neural networks.
problem Inference from time sequences with limited labeled data.
method Combines model-based algorithms and data-driven ML tools for stationary time sequences.
result Learned factor graphs can accurately infer from small training sets.
Improves scalability and robustness of dynamic graph clustering.
problem Scalability and robustness issues in matrix factorization methods for dynamic graphs.
method Temporal separated matrix factorization, bi-clustering regularization, selective embedding updating.
result Demonstrated scalability, robustness, and effectiveness on synthetic and real-world benchmarks.
We solve learning mixtures of graphs from epidemic cascades, establishing conditions and algorithms.
problem Learning the weighted edges of a balanced mixture of two undirected graphs from epidemic cascades.
method Established necessary and sufficient conditions for polynomial-time solvability, provided efficient algorithms with optimal sample complexity.
result First rigorous conditions and algorithms for learning graph mixtures from epidemic cascades.
New method learns DAGs from data without acyclicity constraint.
problem Learning DAGs from data without imposing acyclicity.
method Sparse matrix factorization and ℓ1-penalized optimization. result Empirical success in recovering true graphs and almost-DAG graphs.
Proposes a new tensor factorization model for better link prediction in knowledge graphs.
problem Lack of information in treating missing and non-existing relations equally in tensor factorization models.
method Introduces a binary tensor factorization model with probit link to address the issue.
result Shows improved prediction accuracy and interpretability compared to existing models.
A new framework models and simulates multibody systems using factor graphs.
problem Solving kinematic and dynamic problems for multi-body systems.
method Factor graph theory for modeling and simulation of multibody systems.
result The proposed framework provides a unified approach for multibody systems.
Accelerated Gibbs sampling for Gaussian graphical models using dual factor graphs.
problem Improving convergence rate of Gibbs sampling for Gaussian graphical models.
method Dual normal factor graph approach to accelerate convergence.
result Universal convergence rate improvement in dual domain for all homogeneous models.
GRU-PFG model extracts inter-stock correlations from stock factors using graph neural networks.
problem Limited effectiveness of models relying solely on stock factors for capturing stock correlations.
method Project stock factors into a graph and use graph neural networks to extract inter-stock correlations.
result Achieves better prediction results than models relying solely on stock factors and comparable to second category models.
Paper proposes AI for stock market forecasting using external knowledge.
problem Forecasting stock prices influenced by external factors.
method Learning from historical data and external temporal knowledge graphs modeled as Hawkes processes.
result Dynamic representations effectively rank stocks based on returns.
IPGDN learns disentangled node representations in graphs.
problem Learning disentangled node representations in graph convolutional networks (GCNs).
method IPGDN uses neighborhood routing mechanism and HSIC to enforce independence among latent representations.
result IPGDN outperforms state-of-the-arts in graph classification, clustering, and visualization.
Graph morphism maps Poisson cocycles to symmetries, revealing factorization through Jacobi identity.
problem Mapping graph cocycles to symmetries of Poisson structures.
method Kontsevich graph orientation morphism and differential consequences of Jacobi identity.
result Existence of factorization through differential consequences of Jacobi identity.
SimplE enhances tensor factorization for better link prediction in knowledge graphs.
problem Link prediction in knowledge graphs to discover new relationships.
method Proposes SimplE, a simple enhancement of CP decomposition to learn entity embeddings dependently.
result SimplE outperforms state-of-the-art tensor factorization techniques in link prediction.
Graph neural networks speed up nonnegative matrix factorization.
problem Efficiently factorize nonnegative matrices for various applications.
method Developed a graph neural network that combines bipartite self-attention with ADMM updates.
result Significant acceleration achieved in nonnegative matrix factorization.
New method approximates partition function of graphical models using gauge functions and polynomials.
problem Computing the partition function of graphical models is computationally challenging.
method Combines gauge function technique with real stable polynomials to approximate partition function.
result Belief Propagation estimations in the sequence do not decrease and low-bound the partition function.
Study abelian factors in Lie algebras from graph edge labels.
problem Understanding abelian factors in Lie algebras from graph edge labels.
method Analyzing 2-step nilpotent Lie algebras constructed from graphs, computing abelian factors, and studying singularity properties.
result Explicit computation of abelian factors for various graph families.
Extract common latent factors from graphs for better representation learning.
problem Graph-level representation learning challenges due to limited labeled data and poor negative sample selection.
method Graph-wise Common Latent Factor Extraction (GCFX) using deepGCFX model.
result Improved graph-level and node-level tasks performance compared to state-of-the-art methods.
Novel method models dynamic brain graphs from time series data.
problem Generating hypotheses for dynamic brain states.
method Conditionally weighted superposition of static graphs.
result Improves f1-scores by 22-28% on average over baselines.
This paper defines a duality for labeled graphs and factorizations, linking them to graph embeddings and Hurwitz enumeration.
problem Understanding the relationship between labeled graphs and factorizations in symmetric groups.
method Defining a mind-body duality and interpreting it in terms of Properly Embedded Graphs and Cellularly Embedded Graphs.
result Established a connection between factorizations, labeled graphs, and graph embeddings, including applications to Hurwitz enumeration.
New method for nonlinear filtering and smoothing using factor graphs.
problem Handling deterministic nonlinear transformations in factor graphs.
method Approximate Gaussian message passing rules for factor graphs with Markov property.
result Proposed nonlinear modified Bryson-Frazier smoother.
The free factor graph for Aut(F_N) is not hyperbolic.
problem Characterizing the geometry of the free factor graph for Aut(F_N).
method Analyzing the quasi-isometric embedding of orbits in the graph of free factors.
result The free factor graph for Aut(F_N) is not hyperbolic.
DGA and DVGA learn disentangled graph representations to improve graph analysis.
problem Holistic graph auto-encoders fail to capture latent factors effectively.
method Design disentangled graph convolutional network and component-wise flow, impose independence constraints.
result Improved disentangled graph representations enhance graph analysis tasks.
Data-driven factor graphs improve BCJR detection robustness.
problem Implementing BCJR detection with accurate channel model knowledge.
method Learn factor graph using machine learning from labeled data.
result BCJRNet learns to implement BCJR detection from small training sets.
End-to-end graph-based SSL learns all graph factors dynamically.
problem Learning quality of graph in SSL is crucial but difficult.
method Proposes an end-to-end approach to optimize all graph factors.
result Demonstrates effectiveness on benchmark datasets.
NCFA uses deep learning and causal discovery to analyze complex data.
problem Analyzing complex, interdependent data with causal relationships.
method NCFA combines latent causal discovery and variational autoencoders.
result NCFA outperforms standard VAEs in sparsity, complexity, and causal interpretability.
Paper tackles graph estimation with approximate recovery criteria, matching exact recovery bounds in many cases.
problem Estimating the graph of an Ising model with approximate recovery criteria.
method Adopting approximate recovery criterion, using Fano's inequality and graph ensembles to derive lower bounds.
result Lower bounds on sample complexity match exact recovery bounds in many cases, indicating similar difficulty.
New model clusters graphs using Gromov-Wasserstein discrepancy.
problem Graph clustering with topological structures and node attributes.
method Gromov-Wasserstein discrepancy for relational graph comparison; learns atoms and weights via minimization of discrepancy.
result Model achieves flexible factorization of unaligned graphs with different sizes.
Efficiently learns deep factor graphs using Gaussian belief propagation.
problem Learning in deep factor graphs with efficient inference.
method Treats all relevant quantities as random variables, uses belief propagation for inference.
result Efficiently solves training and prediction problems in deep factor graphs with belief propagation.
We prove a mapping between dual and primal factor graph marginals for efficient estimation.
problem Efficient estimation of marginal densities in factor graphs.
method Local mappings derived from Fourier transforms of local factors, applied to Ising and Potts models.
result Marginal densities can be more accurately estimated in the dual domain.
New method clusters graph topology and attributes efficiently.
problem Clustering attributed graphs with complex topology-attribute relationships.
method Symmetric NMF with PU learning for non-linear projection.
result Outperforms existing methods in clustering quality.
We propose a new yet natural algorithm for learning the graph structure of general discrete graphical models (a.k.a. Markov random fields) from samples. Our algorithm finds the neighborhood of a node by sequentially adding nodes that produce the largest reduction in empirical conditional entropy; it is greedy in the se…
Graphs help agents learn emergent communication.
problem Learning robust communication protocols in multi-agent systems.
method Graph convolutional networks to model emergent language and cooperation.
result Agents learn to generalize beyond training samples, revealing true factors of variation.
The paper introduces a pooling mechanism for graph CNNs using NMF.
problem Pooling in graph structured data for efficient computation.
method Non-negative matrix factorization for node pooling.
result The pooling mechanism improves graph classification performance.
Develops a deep multi-factor model for factor investing with clear financial insights.
problem Lack of interpretability and unclear financial insights in non-linear factor models.
method Industry and market neutralization modules, graph attention modules, factor-attention module.
result Demonstrates effectiveness in factor investing with real-world stock market data.
Graphs model human mobility patterns, reducing errors in data matching.
problem Lack of high-quality data and computational resources for graph-based mobility analysis.
method Embedding graphs into a continuous space to address matching, modeling, and visualization challenges.
result Approx 40% decrease in error on average in matched graphs vs unmatched ones.
The paper refines 2-factor homology to a stable homotopy type for planar trivalent graphs with perfect matchings.
problem Developing a stable homotopy type for planar trivalent graphs with perfect matchings.
method Defining a cover functor from the 2-factor flow category to the cube flow category, realizing the 2-factor spectrum, and showing it's an invariant.
result The stable homotopy type of the 2-factor spectrum is an invariant of planar trivalent graphs with perfect matchings.