Study pairs of subspaces with or without a common complement in Hilbert spaces.
problem Characterize pairs of subspaces with or without a common complement in Hilbert spaces.
method Analyze pairs of subspaces (S, T) in the Grassmann manifold Gr(H) of a Hilbert space H, identifying Delta and Gamma based on the existence of a common complement.
result Delta is open and its connected components are parametrized by dimension and codimension. Gamma is a C^\infty submanifold characterized by dimensions and semi-Fredholm indices.
Common Representation Learning (CRL), wherein different descriptions (or views) of the data are embedded in a common subspace, is receiving a lot of attention recently. Two popular paradigms here are Canonical Correlation Analysis (CCA) based approaches and Autoencoder (AE) based approaches. CCA based approaches learn …
In this work we propose a method for reducing the dimensionality of tensor objects in a binary classification framework. The proposed Common Mode Patterns method takes into consideration the labels' information, and ensures that tensor objects that belong to different classes do not share common features after the redu…
This paper offers a new algebraic perspective of GCCA using subspace intersection.
problem Finding common variables across multiple feature representations.
method Subspace intersection approach based on a (bi-)linear generative model.
result GCCA is equivalent to subspace intersection, with conditions for identifiable common subspace.
Bayesian methods reduce variance in subspace identification for small data sets.
problem High variance in traditional subspace identification methods for large models or small sample sizes.
method Investigation of Bayesian estimation solutions (regularized and shrinkage estimators) for subspace identification.
result Bayesian estimators reduce estimation risk by up to 40% compared to traditional methods.
WPCA improves subspace recovery robustness to outliers.
problem Improving subspace recovery in the presence of outliers.
method Winsorized PCA (WPCA) with theoretical analysis of accuracy and robustness.
result WPCA provides consistent subspace recovery from contaminated data.
The assumption of positivity in causal inference (also known as common support and co-variate overlap) is necessary to obtain valid causal estimates. Therefore, confirming it holds in a given dataset is an important first step of any causal analysis. Most common methods to date are insufficient for discovering non-posi…
Due to the lack of enough generalization in the state-space, common methods in Reinforcement Learning (RL) suffer from slow learning speed especially in the early learning trials. This paper introduces a model-based method in discrete state-spaces for increasing learning speed in terms of required experience (but not r…
We investigate a Gaussian mixture model (GMM) with component means constrained in a pre-selected subspace. Applications to classification and clustering are explored. An EM-type estimation algorithm is derived. We prove that the subspace containing the component means of a GMM with a common covariance matrix also conta…
A method for identifying joint and individual subspaces from multi-view data.
problem Unclear conditions for reliably identifying joint and individual subspaces from noisy, high-dimensional measurements.
method Rigorously quantifies conditions based on signal rank, principal angles, and noise levels. Characterizes spectrum perturbations of product of projection matrices.
result Estimates joint and individual subspaces more accurately than existing approaches in simulations and real-world applications.
Subspace clustering is the unsupervised grouping of points lying near a union of low-dimensional linear subspaces. Algorithms based directly on geometric properties of such data tend to either provide poor empirical performance, lack theoretical guarantees, or depend heavily on their initialization. We present a novel …
This work improves sample efficiency in meta-learning for nonlinear tasks.
problem Learning complex tasks efficiently with limited data.
method Subspace-based representations for nonlinear tasks.
result Subspace-based representations can be learned efficiently and improve future task performance.
In this paper, we propose a novel method for projecting data from multiple modalities to a new subspace optimized for one-class classification. The proposed method iteratively transforms the data from the original feature space of each modality to a new common feature space along with finding a joint compact descriptio…
It is often the case that, within an online recommender system, multiple users share a common account. Can such shared accounts be identified solely on the basis of the userprovided ratings? Once a shared account is identified, can the different users sharing it be identified as well? Whenever such user identification …
A new method tackles bilevel optimization using Lanczos process for efficient hyper-gradient computation.
problem Efficiently solving large-scale bilevel optimization problems with gradient-based methods.
method Constructing low-dimensional approximate Krylov subspaces with the Lanczos process to approximate the Hessian inverse vector product.
result Demonstrates a O(ε−1) convergence rate and efficiency in synthetic and deep learning tasks. We introduce a two step algorithm with theoretical guarantees to recover a jointly sparse and low-rank matrix from undersampled measurements of its columns. The algorithm first estimates the row subspace of the matrix using a set of common measurements of the columns. In the second step, the subspace aware recovery of …
This paper considers the problem of clustering a collection of unlabeled data points assumed to lie near a union of lower-dimensional planes. As is common in computer vision or unsupervised learning applications, we do not know in advance how many subspaces there are nor do we have any information about their dimension…
Given an overcomplete dictionary A and a signal b=Ac∗ for some sparse vector c∗ whose nonzero entries correspond to linearly independent columns of A, classical sparse signal recovery theory considers the problem of whether c∗ can be recovered as the unique sparsest solution to b=Ac. It is now well-…
New method disentangles latent subspaces under correlation shifts.
problem Correlations between factors of variation make disentanglement models less robust.
method Enforces independence between subspaces conditioned on available attributes using adversarial CMI minimization.
result Models are disentangled and robust under correlation shifts, including in weakly supervised settings.
The multimodal web elements such as text and images are associated with inherent memory costs to store and transfer over the Internet. With the limited network connectivity in developing countries, webpage rendering gets delayed in the presence of high-memory demanding elements such as images (relative to text). To ove…
Study non-asymptotic bounds on correlation in high-dimensional linear systems, revealing invariant subspaces and bottlenecks.
problem Understanding correlation and mixing in high-dimensional linear systems with Gaussian noise.
method Sampling from sub-trajectories, using Talagrand's inequality, and analyzing invariant subspaces.
result Large discrepancy between algebraic and geometric multiplicity leads to bottlenecks between invariant subspaces.
Algorithm removes spurious concepts from neural network representations without harming task performance.
problem Spurious correlations hinder neural network out-of-distribution generalization.
method Iterative algorithm that identifies two orthogonal subspaces in neural network representation.
result Algorithm outperforms existing methods on computer vision and natural language processing benchmarks.
The paper proposes methods to find a shared active subspace for multivariate vector-valued functions.
problem Minimizing the deviation between function evaluations in the original and reconstructed spaces.
method Manipulating gradients or SPD matrices to identify a shared structure.
result Summing SPD matrices often identifies the best shared active subspace.
Multi-view subspace learning (MSL) aims to find a low-dimensional subspace of the data obtained from multiple views. Different from single view case, MSL should take both common and specific knowledge among different views into consideration. To enhance the robustness of model, the complexity, non-consistency and simil…
A new machine learning methodology, called successive subspace learning (SSL), is introduced in this work. SSL contains four key ingredients: 1) successive near-to-far neighborhood expansion; 2) unsupervised dimension reduction via subspace approximation; 3) supervised dimension reduction via label-assisted regression …
PAS method improves UDA by progressively refining subspaces for reliable pseudo-labels.
problem Mitigating mode collapse in unsupervised domain adaptation.
method Progressive Adaptation of Subspaces (PAS) approach.
result PAS effectively mitigates mode collapse and improves performance in UDA and PDA.
Study optimizes shared singular subspace estimation from noisy matrices.
problem Estimating shared singular subspaces across multiple noisy matrices.
method Low-rank matrix denoising framework with Stack-SVD and novel estimators.
result Stack-SVD achieves minimax rate-optimality for identical shared subspaces, and novel estimators for partial sharing.
New subspace prototype flag median improves clustering on noisy data.
problem Finding robust prototypes for datasets of images and videos.
method Proposes flag median and introduces FlagIRLS algorithm for its calculation.
result Flag median is robust to outliers and improves cluster purity.
In a recent paper Baker and Bowler introduced matroids over hyperfields, offering a common generalization of matroids, oriented matroids, and linear subspaces of based vector spaces. This paper introduces the notion of a topological hyperfield and explores the generalization of Grassmannians and realization spaces to t…
Gradient-based meta-learning methods leverage gradient descent to learn the commonalities among various tasks. While previous such methods have been successful in meta-learning tasks, they resort to simple gradient descent during meta-testing. Our primary contribution is the {\em MT-net}, which enables the meta-learner…
Novel approach embeds loss tunnels in neural networks, revealing insights into their structure.
problem Understanding the structure of neural network loss surfaces, especially low-loss tunnels.
method Directly embedding loss tunnels into the loss landscape of neural networks.
result Improved insights into the length and structure of loss tunnels, and better subspace inference in Bayesian neural networks.
A new method compresses NLP networks by using multiple subspaces instead of a single one.
problem Large errors in compressing NLP networks using a single subspace.
method Projective clustering to find multiple subspaces that minimize squared distances.
result Networks are more accurate and smaller compared to standard matrix factorization.
LOL method simplifies forming linear combinations of latent variables.
problem Lack of general-purpose methods for manipulating latent variables.
method Latent Optimal Linear combinations (LOL) method.
result LOL simplifies creation of expressive low-dimensional representations.
Relationships between entities in datasets are often of multiple nature, like geographical distance, social relationships, or common interests among people in a social network, for example. This information can naturally be modeled by a set of weighted and undirected graphs that form a global multilayer graph, where th…
Proposes HeteroJIVE for joint subspace estimation in multi-view data with statistical and structural heterogeneity.
problem Joint subspace estimation in multi-view data with varying statistical and structural heterogeneity.
method HeteroJIVE: A weighted two-stage spectral algorithm addressing statistical and structural heterogeneity.
result HeteroJIVE achieves the O(K−1/2) rate without iterative refinement, validating the oracle-optimal weighting scheme. The paper analyzes PLS-SVD in high-dimensional data integration, revealing its strengths and limitations.
problem Understanding the behavior of PLS-SVD in high-dimensional data integration.
method Analysis using random matrix theory and singular value decomposition.
result PLS-SVD exhibits counter-intuitive or limiting behavior in certain regimes and outperforms PCA when detecting common latent subspace.
We will develop simple relations between the arc-lengths of a pair of geodesics that share common end-points. The two geodesics differ only by the requirement that one is constrained to lie in a subspace of the parent manifold. We will present two applications of our results. In the first example we explore the converg…
Over the past few decades, we have witnessed a large family of algorithms that have been designed to provide different solutions to the problem of dimensionality reduction (DR). The DR is an essential tool to excavate the important information from the high-dimensional data by mapping the data to a low-dimensional subs…
A framework for federated learning with heterogeneous data.
problem Federated learning with data from clients using different data representations.
method FLIC framework that maps client data into a common feature space via local embedding functions, learned federally using Wasserstein barycenters and trained locally via distribution alignment.
result FLIC outperforms FL benchmarks with heterogeneous input feature spaces.
In generalized complex geometry, we revisit linear subspaces and submanifolds that have an induced generalized complex structure. We give an expression of the induced structure that allows us to deduce a smoothness criteria, we dualize the results to submersions and we make a few comments on generalized complex mapping…
Unified PCA framework on flag manifolds for robust data analysis.
problem Outliers and manifold data in PCA.
method Generalization of PCA to flag manifolds, optimization problems, and tangent-PCA integration.
result Novel robust and dual geodesic PCA variations.
Modern machine learning algorithms have been adopted in a range of signal-processing applications spanning computer vision, natural language processing, and artificial intelligence. Many relevant problems involve subspace-structured features, orthogonality constrained or low-rank constrained objective functions, or sub…
We propose a new learning-based approach to solve ill-posed inverse problems in imaging. We address the case where ground truth training samples are rare and the problem is severely ill-posed - both because of the underlying physics and because we can only get few measurements. This setting is common in geophysical ima…
Proposes a method for multi-view clustering that considers local structures and feature weights.
problem Challenges in effectively exploiting complementary information across multiple views.
method Simultaneously assigns weights to different features and captures local information in view-specific feature spaces.
result Achieves state-of-the-art performance on benchmark datasets.
Proposes KMvDA for object recognition from multi-view data.
problem Recognizing objects from different views, even when views are heterogeneous.
method Introduces kernel multi-view discriminant analysis (KMvDA) and uses random Fourier features (RFF) for large-scale learning.
result KMvDA and RFF approximation improve object recognition from multi-view data.
New method clusters multi-view data by squeezing hybrid knowledge.
problem Removal of redundant information and fusion of multi-view features.
method Low-rank subspace multi-view clustering with adaptive graph regularization.
result Our method outperforms state-of-the-art algorithms on multi-view benchmarks.
The paper addresses data uncertainty in graph embedding by modeling data points as Gaussian distributions.
problem Data uncertainty in machine learning pipelines leads to misleading embeddings and lower accuracy.
method The paper proposes modeling data uncertainty using Gaussian distributions and reformulates graph embedding techniques.
result The proposed methods improve the accuracy of graph embedding by accounting for data uncertainty.
A flag is a sequence of nested subspaces. Flags are ubiquitous in numerical analysis, arising in finite elements, multigrid, spectral, and pseudospectral methods for numerical PDE; they arise in the form of Krylov subspaces in matrix computations, and as multiresolution analysis in wavelets constructions. They are comm…