Paper recovers multi-subspace matrices from permuted data.
problem Recovering a multi-subspace matrix from permuted data with corrupted columns.
method Four-stage algorithm pipeline: outlier identification, subspace reconstruction, outlier classification, unsupervised sensing.
result The pipeline provides theoretical guarantees for reliable multi-subspace matrix recovery.
This paper tackles permutation recovery in unlabeled sensing from multiple measurement vectors.
problem Permutation recovery in unlabeled sensing from multiple measurement vectors.
method The paper studies the case of multiple noisy measurement vectors (MMVs) resulting from a common permutation and proposes computational schemes for permutation recovery.
result A large stable rank of the signal significantly reduces the required signal-to-noise ratio (SNR) for permutation recovery, and the problem can be solved efficiently using ADMM.
Recently, the method of b-bit minwise hashing has been applied to large-scale linear learning and sublinear time near-neighbor search. The major drawback of minwise hashing is the expensive preprocessing cost, as the method requires applying (e.g.,) k=200 to 500 permutations on the data. The testing time can also be ex…
C-MinHash reduces the number of permutations needed for MinHash from thousands to just two.
problem Approximating Jaccard similarity in large binary datasets using many permutations.
method Initial permutation followed by circulant shifting of a second permutation to generate hashes.
result C-MinHash achieves unbiased Jaccard similarity estimation with uniformly smaller variance.
A new method uses vectorized summaries of persistence diagrams for efficient hypothesis testing.
problem Efficient hypothesis testing for large and complex persistence diagrams.
method Vectorized summaries of Betti functions and a new shuffling technique.
result The vectorized Betti function leads to competitive results compared to baseline methods.
The paper models financial correlation matrices using permutation invariant Gaussian models and predicts market anomalies.
problem Modeling and predicting financial correlation matrices from high-frequency data.
method Constructing permutation invariant Gaussian matrix models with 4 parameters, using graph theory and polynomial functions.
result The permutation invariant Gaussian matrix model predicts the expectation values of cubic and quartic polynomials with strong evidence of fit.
This paper compresses neural networks by permuting and quantizing weights.
problem Efficiently compressing large neural networks for resource-constrained platforms.
method Permuting and quantizing weights, connecting to rate-distortion theory, and using annealed quantization.
result Significant compression with minimal accuracy loss, e.g., 40-70% reduction in gap with uncompressed model.
OPORP combines permutation and random projection for efficient data vector compression.
problem Efficiently estimating cosine similarity in embedding-based retrieval applications.
method OPORP uses a permutation followed by a random vector dot product, then aggregates and normalizes the results into bins.
result OPORP improves the estimation of cosine similarity, reducing variance and improving accuracy.
Derives formulae for general permutation equivariant layers and presents a second order graph variational encoder.
problem Tackles the limitation of previous equivariant neural networks by considering permutations of matrices.
method Derives formulae for general permutation equivariant layers, including matrix permutations. Presents a second order graph variational encoder.
result Latent distribution of equivariant generative models must be exchangeable.
π-GNN learns soft permutations for graph representations, improving graph classification and regression.
problem Limitations of MPNNs in graph neural networks.
method Proposes π-GNN, which learns a soft permutation matrix for each graph, projecting graphs into a common vector space.
result π-GNN achieves performance competitive with state-of-the-art models on graph classification and regression tasks.
Paper presents a novel neural network for MIMO symbol detection.
problem Handling a variable number of users in MIMO systems.
method Recurrent and permutation equivariant neural network architecture with iterative decoding.
result The neural detector outperforms existing methods in accuracy and efficiency.
C-OPH improves One Permutation Hashing by using a shorter circulant permutation.
problem Improving the accuracy of One Permutation Hashing (OPH) for Jaccard similarity estimation.
method Develops a new densification method using a shorter circulant permutation.
result Achieves the smallest estimation variance for Jaccard similarity.
Consider a noisy linear observation model with an unknown permutation, based on observing y=Π∗Ax∗+w, where x∗∈Rd is an unknown vector, Π∗ is an unknown n×n permutation matrix, and w∈Rn is additive Gaussian noise. We analyze the problem of permutation recovery in a …
Paper proposes a differentially private test for joint dependence among random vectors.
problem Detecting joint dependence among sensitive data while maintaining privacy.
method Differentially private permutation methodology for dHSIC test.
result Proposed test attains minimax optimal power across privacy regimes.
Study recovers spike order in noisy tensor estimation without SNR assumptions.
problem Estimating multiple signal vectors from noisy tensor observations.
method Gradient flow optimization of a nonconvex function.
result Determines sample complexity for efficient permutation recovery.
ExNODE uses ODE to model sets with permutation equivariance.
problem Capturing intra-set dependencies in unordered sets.
method Exchangeable Neural ODE (ExNODE) using ODE.
result ExNODE achieves permutation equivariance for set modeling.
A new loss function for set reconstruction without order consideration.
problem Reconstructing sets of elements without considering their order.
method Set Cross Entropy, a permutation-invariant loss function.
result Natural information-theoretic interpretation and successful evaluations in tasks.
Deep networks exhibit permutation saddles and valleys between equivalent minima.
problem Understanding the structure of loss landscapes in deep neural networks.
method Geometric approach to constructing paths between equivalent minima and saddle points.
result Existence of permutation saddles and valleys in deep neural networks.
We study the problem of designing models for machine learning tasks defined on \emph{sets}. In contrast to traditional approach of operating on fixed dimensional vectors, we consider objective functions defined on sets that are invariant to permutations. Such problems are widespread, ranging from estimation of populati…
Signed-permutation coordinate transport improves model alignment across checkpoints.
problem Improper alignment of coordinate-indexed objects across model checkpoints.
method Introduces sign-marginalized Hungarian matching and coordinate-preserving transport.
result Recovering signed-permutation gauge improves coordinate alignment and model performance.
A structured prediction method for ranking labels.
problem Solving label ranking problems as structured output regression.
method Two-step approach: regression in feature space followed by pre-image solving.
result Efficiency on real-world datasets for partial and complete rankings.
Algorithm recovers permutations of high-dimensional Gaussian vectors with constant correlation.
problem Recovering permutations of high-dimensional Gaussian vectors with constant correlation.
method Computing and comparing weighted counts of specially chosen wide trees.
result Polynomial-time algorithm for exact recovery at constant correlation.
New basis for permutation equivariant layers reduces computation costs.
problem Efficiently computing permutation equivariant layers in neural networks.
method Generalized partition algebra basis with low-rank tensors.
result Low-rank tensors enable faster computation compared to orbit basis.
Proposes deep learning methods for handling random vectors.
problem Handling flexible input data like probability measures in deep learning.
method Develops deep architectures to handle permutation invariances, varying weights, and cardinality.
result Demonstrates the effectiveness of deep architectures on measures for classification, reduction, and prediction.
We propose an algorithm to separate simultaneously speaking persons from each other, the "cocktail party problem", using a single microphone. Our approach involves a deep recurrent neural networks regression to a vector space that is descriptive of independent speakers. Such a vector space can embed empirically determi…
To model categorical response variables given their covariates, we propose a permuted and augmented stick-breaking (paSB) construction that one-to-one maps the observed categories to randomly permuted latent sticks. This new construction transforms multinomial regression into regression analysis of stick-specific binar…
Paper connects GLM and LRM for better classification performance.
problem Improving classification performance using statistical inference.
method Derives a statistical test based on SVM and permutation analysis.
result MLE-based inference provides better parameter estimation.
Study multiplicity-free covering of graded manifolds, proving equivalence of categories.
problem Equivalence of categories of graded manifolds and symmetric vector bundles.
method Defined and computed multiplicity-free covering, showed deck transformation group isomorphic to Sn. result Equivalence of categories of graded manifolds and symmetric n-fold vector bundles. Matching correlated VAR time series databases by recovering matching permutations.
problem Matching perturbed and permuted correlated VAR time series.
method Probabilistic framework modeling, maximum likelihood estimator (MLE), linear assignment, convex relaxations.
result Recovery guarantees for perfect or partial recovery of matching permutations, thresholds for σ. Let M be a smooth (C∞) manifold, F1,...,Fn be vector fields on M generating the corresponding flows Φ1,...,Φn, and α1,...,αn:M→R smooth functions. Define the following map f:M→M by f(x)=Φn(...(Φ2(Φ1(x,α1(x)),α2(x)),...,αn(x)). In this note we give a necessa…
Bayesian optimization for set inputs using approximate set kernels.
problem Permutation-invariant optimization over sets with black-box functions.
method Developed a Bayesian optimization method with set kernel, efficient approximate set kernel, and constrained acquisition function.
result Our method outperforms other methods in numerical experiments.
PINE embeds graph nodes flexibly, capturing any neighbor dependency.
problem Learning flexible node representations from graph neighborhoods.
method PINE uses partial permutation invariant set functions to capture any possible neighbor dependencies.
result PINE outperforms state-of-the-art methods on various graph learning tasks.
Model learns set representations through optimized permutations.
problem Challenges in learning set representations due to permutation-invariance.
method Proposes a Permutation-Optimisation module to learn set permutations.
result Achieves state-of-the-art results on various set learning tasks.
Cheap permutation tests speed up distribution testing without sacrificing accuracy.
problem Efficiently testing distribution differences and independence.
method Group datapoints into bins and permute only these bins, using stored sufficient statistics.
result Cheap permutation tests maintain the accuracy and optimality of standard tests but are significantly faster.
Random permutations can offer faster convergence than with-replacement sampling for some functions.
problem Understanding when and how random permutations outperform with-replacement sampling in SGD convergence.
method Analyzing convergence rates for different function classes (1D strongly convex, general strongly convex, quadratic strongly convex).
result The optimal convergence gap between random and permutation-based SGD varies from exponential to nonexistent, depending on the function class.
A new method for generating sets and graphs without requiring exchangeability.
problem Generating exchangeable distributions for sets and graphs is challenging.
method Top-n creation, a differentiable generation mechanism that selects relevant points from a latent vector.
result Top-n method outperforms i.i.d. generation in various tasks.
Permutations linked to knots and links, with unknots counted by Schröder numbers.
problem Understanding permutations as knots and links.
method Using grid diagrams and Bennequin's inequality.
result Permutations corresponding to unknots and links are counted by Schröder numbers.
This study compares feature importance and explainability in quantum vs classical ML models.
problem Lack of transparency in ML models, especially in sensitive fields.
method Comparison of classical ML (SVM, Random Forest) and hybrid quantum ML (VQC, QSVC) models using feature importance and explainability methods.
result Quantum ML models provide insights similar to classical models but with unique quantum features.
We tackle permutation in linear regression with a new inference framework.
problem Statistical investigation of permutation in linear regression models.
method Localization step followed by conditional Monte Carlo test and coefficient inference.
result Valid statistical inference procedures for permutation and regression coefficients.
Regularizes RNNs to be invariant to input order.
problem Making RNNs invariant to input order.
method Stochastic regularization to enforce permutation invariance.
result Improves model performance on permutation invariant tasks.
Permutability of surface transforms yields discrete analogs.
problem Discretization of smooth surfaces with specific properties.
method Permutability of transforms of smooth surfaces.
result Discrete surfaces with discrete analogs of original properties.
Janossy pooling averages permutation-sensitive functions over all sequences to create invariant functions.
problem Creating deep, invariant functions for variable-size inputs.
method Janossy pooling: average permutation-sensitive functions over all reorderings.
result Improved performance over state-of-the-art methods.
AutoShuffleNet learns permutation matrices in CNNs for improved accuracy.
problem Manual design of channel shuffling in ShuffleNet.
method Learning permutation matrices via an exact Lipschitz continuous penalty in deep learning.
result Improved classification accuracies on CIFAR-10 and ImageNet datasets.
New equivalences found between graded supermanifolds and vector bundles.
problem Understanding equivalences between graded supermanifolds and vector bundles.
method Explicit geometric constructions using supergeometry tools.
result Desuperization equivalence functor as a composition of canonical equivalences.
Dynamic systems linked to infinite permutation matrices.
problem Dynamic equivalence of control systems.
method Association of infinite permutation matrices.
result Relationship between dynamic equivalences and permutation matrices.
A new permutation method improves two-sample testing power.
problem Two-sample testing with improved power and validity.
method Structured block-restricted cross-swaps.
result Block-restricted permutations achieve higher power than full permutations.
Survey on neural networks for set-based data.
problem Efficient processing of set-based inputs in machine learning.
method Exploration of Deep Sets and Transformers for set functions approximation.
result Deep Sets can be generalized by differences in aggregation function.
New method for decomposing vectors into independent components over finite alphabets.
problem Decomposing vectors into independent components without prior generation assumptions.
method Branch and bound search tree algorithm, linear approximations, order permutation.
result Efficiently decomposes the majority of vectors into independent components.