Polytopic Matrix Factorization models data as latent vectors from a polytope, maximizing determinant for identifiability.
problem Data decomposition with semi-structured latent vectors and polytope constraints.
method Model input data as latent vectors from a polytope, using determinant maximization for identifiability.
result Identifiability condition for polytopes with specific symmetry restrictions.
Proposes D-CDLF for multi-view data decomposition.
problem Uncorrelatedness between common and distinctive latent factors.
method Decomposes data into common, distinctive, and noise components.
result Effective uncorrelatedness between distinctive latent factors from different views.
We consider analysis of relational data (a matrix), in which the rows correspond to subjects (e.g., people) and the columns correspond to attributes. The elements of the matrix may be a mix of real and categorical. Each subject and attribute is characterized by a latent binary feature vector, and an inferred matrix map…
Enhances matrix completion with pairwise penalties for latent features.
problem Improving prediction performance in matrix completion.
method Proposes a general optimization framework with non-/convex pairwise penalty functions and develops an efficient algorithm.
result The proposed framework outperforms standard matrix completion methods, especially in scenarios with latent subgroup structures.
New test for latent block models to determine cluster numbers.
problem No statistical test for latent block models.
method Developed a goodness-of-fit test using random matrix theory.
result Demonstrated the effectiveness of the test method.
A new model Weighted-SVD improves recommendation accuracy by adjusting latent factor weights.
problem Current Matrix Factorization models assume equal weights for all latent factors, which may not be accurate.
method Integrates linear regression with SVD to allow different weights for latent factors.
result The Weighted-SVD model outperforms other models in RMSE metrics on multiple datasets.
New criterion ensures recovery of latent factors in NMF with mild conditions.
problem Identifying latent factors in nonnegative matrix factorization (NMF) under mild conditions.
method Proposed a new identification criterion based on the scatteredness of one factor's rows in the nonnegative orthant.
result Latent factors can be provably identified from the NMF model with minimal structural assumptions.
GAME improves matrix completion by considering subgroup-specific latent structures.
problem Heterogeneous data with overlapping categories, smoothing away subgroup-specific variation.
method Group-Aware Matrix Estimation (GAME) with overlapping nuclear-norm penalties.
result GAME outperforms global low-rank estimators in structured missingness regimes.
Improves matrix completion by exploiting biased observation patterns.
problem Matrix completion with biased observation patterns.
method Mask Nearest Neighbor (MNN) algorithm: two-stage process.
result MNN achieves competitive performance with 28x smaller mean squared error.
We consider the problem of covariance matrix estimation in the presence of latent variables. Under suitable conditions, it is possible to learn the marginal covariance matrix of the observed variables via a tractable convex program, where the concentration matrix of the observed variables is decomposed into a sparse ma…
An ADRC-incorporated SGD algorithm improves latent factor analysis speed and accuracy.
problem Slow convergence in standard SGD for HDI matrix analysis.
method Incorporates ADRC principles to refine historical and future learning error states.
result Empirically outperforms state-of-the-art LFA models in HDI matrix prediction.
Estimates latent inner products from an anisotropic Gaussian graph with improved spectral method.
problem Recovering latent inner products from an anisotropic Gaussian random geometric graph.
method Doubly centered adjacency matrix, rank-d spectral approximation, Hermite expansion, decoupling argument.
result Estimator achieves mean squared error rate matching state of the art for isotropic case and ill-conditioned covariance matrices.
Data often comes in the form of an array or matrix. Matrix factorization techniques attempt to recover missing or corrupted entries by assuming that the matrix can be written as the product of two low-rank matrices. In other words, matrix factorization approximates the entries of the matrix by a simple, fixed function-…
NMF with specific constraints is equivalent to LDA.
problem Dimensionality reduction of non-negative data.
method NMF with ℓ1 normalization constraints and Dirichlet prior. result NMF with these constraints is equivalent to LDA.
New deep learning model for matrix completion combining linear and nonlinear relationships.
problem Matrix completion considering only linear or nonlinear relations, ignoring latent relationships.
method Combines linear and nonlinear models in a latent variables framework, using a deep neural network with two branches for columns and rows, and manifold learning as an auxiliary task.
result Experimental results show the proposed method outperforms state-of-the-art matrix completion methods.
Paper proposes an algorithm for automatically selecting latent dimensions in NMF.
problem Automatic model selection for NMF with theoretical guarantees.
method Empirical second-order moment and support union recovery.
result The algorithm provably detects the true latent dimensionality.
Simpler method for separating and manipulating latent attributes in autoencoders.
problem Separating and manipulating latent attributes in autoencoders.
method Matrix subspace projection
result Our method allows for changing selected attributes while preserving other information.
GLFA improves latent factor analysis by incorporating graph structures for HiDS matrices.
problem Accurate representation learning on high-dimensional and sparse matrices.
method GLFA incorporates a graph to identify hidden high-order interactions and uses a recurrent LFA structure to improve representation learning.
result GLFA outperforms state-of-the-art models in predicting missing data of HiDS matrices.
Bayesian NMF model improves predictions and avoids overfitting.
problem Predicting missing values and finding hidden patterns in nonnegative data.
method Flexible and hierarchical prior for Bayesian NMF with Gibbs sampling.
result The proposed model leads to better predictions and avoids overfitting.
I-BBS identifies latent sub-manifolds from distance matrices, robust to noise.
problem Identifying latent sub-manifolds from distance matrices in high-dimensional spaces.
method Coordinate-free inference using random distance matrix theory and generative noise models.
result Recovering latent geometry from integer-stable signatures of eigenvalues.
Scalable Gaussian processes with latent Kronecker structure for large datasets.
problem Limited scalability of Gaussian processes for large datasets.
method Leveraging latent Kronecker structure, projecting kernel matrix onto latent Kronecker product, using iterative linear system solvers and pathwise conditioning.
result Outperforms state-of-the-art sparse and variational GPs on real-world datasets with up to five million examples.
We are concerned with an approximation problem for a symmetric positive semidefinite matrix due to motivation from a class of nonlinear machine learning methods. We discuss an approximation approach that we call {matrix ridge approximation}. In particular, we define the matrix ridge approximation as an incomplete matri…
This paper addresses identifiability issues in HLAMs with hierarchical constraints.
problem Identifiability of HLAMs with hierarchical constraints.
method Developed sufficient and necessary identifiability conditions.
result Characterizes impacts of different attribute types in the graph on identifiability.
The paper proposes methods for predicting missing values in mixed data matrices.
problem Matrix completion for mixed data types (continuous, binary, ordinal).
method Generalized latent factor models for low-rank matrix estimation with entrywise consistency.
result Tight probabilistic error bounds for the proposed estimators.
Paper introduces OMD for ordered state transitions in SSMs.
problem Modeling ordered latent states in dynamic systems.
method Ordered Matrix Dirichlet (OMD) prior over ordered stochastic matrices.
result OMD models recover interpretable ordered latent structure without sacrificing predictive performance.
Polynomial-time algorithm learns latent-state systems without spectral radius assumptions.
problem Learning latent-state linear dynamical systems without spectral radius assumptions.
method Spectral filtering technique with a novel convex relaxation.
result Efficient identification of phases for general transition matrices.
New algorithms for latent class analysis using regularized spectral clustering.
problem Identifying latent classes within populations from categorical data.
method Developed two new algorithms using a regularized Laplacian matrix to estimate latent classes.
result Our algorithms provide consistent latent class analysis under mild conditions and can accurately infer the number of latent classes.
New model for high rank matrix completion with online and batch methods.
problem Matrix completion for high rank matrices with latent structure.
method Kernel trick to map data into a high dimensional feature space, explicit parametrization of low dimensional subspace, online fitting procedure.
result Online method can handle streaming data and adapt to non-stationary latent structure.
The paper tackles causal disentanglement with linear models and interventions.
problem Identify latent variables in a causal model from observed data.
method Use linear transformations and interventions to uniquely identify latent variables.
result A single intervention on each latent variable is sufficient for identifying the latent causal model.
Bayesian non-linear matrix completion tackles large, sparse data.
problem Predict missing elements in large, sparsely observed matrices.
method Bayesian Gaussian process latent variable models with data-parallel distributed computation.
result Scalable Bayesian non-linear matrix completion outperforms linear methods.
We study the problem of learning the support of transition matrix between random processes in a Vector Autoregressive (VAR) model from samples when a subset of the processes are latent. It is well known that ignoring the effect of the latent processes may lead to very different estimates of the influences among observe…
Paper proposes a fast algorithm to recover causal DAGs with latent variables.
problem Discovering causal relationships in the presence of latent variables.
method Cholesky factorization of covariance matrix with optimization for latent variables.
result The algorithm significantly outperforms previous methods in synthetic and real-world datasets.
DaConA improves recommendation accuracy with auxiliary data by adapting to different data contexts.
problem Improving recommendation accuracy with auxiliary data considering different data contexts.
method Data context adaptation layer, latent interaction vector, latent independence vector, non-linear function.
result DaConA achieves state-of-the-art accuracy on real-world datasets.
Exact inference possible without observing latent variables in unknown domains.
problem Can exact inference be done without knowing latent variables or their domain?
method Semidefinite programming (SDP) approach based on Karush-Kuhn-Tucker (KKT) conditions and matrix spectrum.
result Exact inference can be achieved without knowing latent variables or their domain.
Proposes a multilayer nonlinear semi-nonnegative matrix factorization for better recommendation.
problem Inaccurate user-item interaction modeling with classical matrix factorization.
method Multilayer nonlinear Semi-NMF approach for latent user and item representations.
result Proposed method achieves better generalization in prediction and comparable representation in clustering.
We face network data from various sources, such as protein interactions and online social networks. A critical problem is to model network interactions and identify latent groups of network nodes. This problem is challenging due to many reasons. For example, the network nodes are interdependent instead of independent o…
Identifies latent actions and dynamics from offline data with diverse demonstrators.
problem Recovering latent actions and environment dynamics from action-free trajectories.
method Assumes distinct policies for each demonstrator, identifies latent transitions and policies via matrix factorization.
result Identifies latent transitions and demonstrator policies up to permutation.
Active seriation recovers item order from noisy pairwise similarity measurements.
problem Recovering an unknown item ordering from noisy pairwise similarity measurements.
method Proposes an active seriation algorithm that provably recovers the latent ordering with high probability.
result Establishes optimal performance guarantees for successful recovery under a uniform separation condition.
Bayesian model fuses diverse microbiome data types.
problem Challenges in fusing different types of microbiome data.
method Flexible multinomial-Gaussian generative model with variational EM algorithm.
result Inferred latent variables provide common dimensionality reduction and predictive posterior distribution.
CDPA identifies common and distinctive patterns in high-dimensional datasets.
problem Existing methods fail to capture the common pattern between coefficient matrices of shared latent factors.
method Proposes CDPA, an unsupervised learning method that incorporates both common and distinctive patterns of coefficient matrices.
result CDPA provides better characterization of common and distinctive patterns in high-dimensional datasets.
Matrices of (approximate) low rank are pervasive in data science, appearing in recommender systems, movie preferences, topic models, medical records, and genomics. While there is a vast literature on how to exploit low rank structure in these datasets, there is less attention on explaining why the low rank structure ap…
We prove a central limit theorem for the components of the largest eigenvectors of the adjacency matrix of a finite-dimensional random dot product graph whose true latent positions are unknown. In particular, we follow the methodology outlined in \citet{sussman2012universally} to construct consistent estimates for the …
Paper shows LDA and SMF have similar generalization errors.
problem LDA and SMF's generalization performance is unknown.
method Algebraic and geometric method to show equivalence of LDA and SMF.
result LDA and SMF have asymptotically same Bayesian generalization error.
Correlation matrices play a key role in many multivariate methods (e.g., graphical model estimation and factor analysis). The current state-of-the-art in estimating large correlation matrices focuses on the use of Pearson's sample correlation matrix. Although Pearson's sample correlation matrix enjoys various good prop…
VAE enhances NMF for probabilistic non-negative matrix factorisation.
problem Non-negative matrix factorisation with probabilistic coefficients.
method Design a VAE network with non-negative weights and non-negative Weibull distribution.
result Effective probabilistic NMF for generating new data and linking latent and input variables.
Algorithm learns latent simplex from perturbed points in input-sparsity time.
problem Learning a latent k-vertex simplex from noisy data. method Input-sparsity time algorithm using low-rank approximation and adaptive selection.
result Algorithm achieves O(extrmnnz(A)) time complexity, avoiding k⋅extrmnnz(A). Hierarchical parametric models consisting of observable and latent variables are widely used for unsupervised learning tasks. For example, a mixture model is a representative hierarchical model for clustering. From the statistical point of view, the models can be regular or singular due to the distribution of data. In …
New method clusters matrix-valued data by latent variables.
problem Clustering matrix-valued data with hidden structure.
method Latent variable model with hierarchical clustering.
result Algorithm attains clustering consistency in high dimensions.