New method for initializing low-rank neural networks improves performance.
problem Training low-rank neural networks efficiently and accurately.
method Inspired by function approximation, proposes a novel low-rank initialization framework.
result Demonstrates significant gap between spectral and low-rank initialization approaches.
We accelerate the power method for strong low-rank approximation using fast sketching.
problem Efficiency bottleneck in power method for large target ranks.
method Developed an algorithmic and theoretical framework for accelerating the power method using fast sketching.
result Simple and provably efficient methods for singular value decomposition, low-rank factorization, and Nyström approximation.
New methods recover best rank-r approximations from few entries.
problem Recovering best rank-r approximations from limited data entries.
method Two agnostic approaches: spectral truncation and projected gradient descent.
result Projected gradient descent yields superior performance.
This paper proposes a new Nystrom-based clustering algorithm for large-scale data.
problem Spectral clustering's high computational complexity for large-scale data.
method Centroid Minimum Sum of Squared Similarities (CMS3) sampling procedure with eigen spectrum shape heuristic.
result Competitive low-rank approximations in test datasets compared to state-of-the-art methods.
Study uses random matrix theory to improve tensor approximation accuracy.
problem Improving tensor approximation accuracy in the presence of noise.
method Random matrix theory applied to tensor unfoldings.
result Characterizes spectral behavior of tensor unfoldings and predicts reconstruction performance.
Given a graphical model (GM), computing its partition function is the most essential inference task, but it is computationally intractable in general. To address the issue, iterative approximation algorithms exploring certain local structure/consistency of GM have been investigated as popular choices in practice. Howev…
In this work, we propose a new randomized algorithm for computing a low-rank approximation to a given matrix. Taking an approach different from existing literature, our method first involves a specific biased sampling, with an element being chosen based on the leverage scores of its row and column, and then involves we…
We propose a new framework for the analysis of low-rank tensors which lies at the intersection of spectral graph theory and signal processing. As a first step, we present a new graph based low-rank decomposition which approximates the classical low-rank SVD for matrices and multi-linear SVD for tensors. Then, building …
Gradient descent with small random init mimics spectral methods for low-rank matrix recovery.
problem Reconstructing a low-rank matrix from few measurements.
method Gradient descent with small random initialization followed by a few iterations.
result Gradient descent from small random init converges to a well-generalizing solution.
Consider the problem of estimating a low-rank matrix when its entries are perturbed by Gaussian noise. If the empirical distribution of the entries of the spikes is known, optimal estimators that exploit this knowledge can substantially outperform simple spectral approaches. Recent work characterizes the asymptotic acc…
Develops precise expressions for random projections for better machine learning tasks.
problem Improving the accuracy of dimensionality reduction in machine learning tasks.
method Exploits recent developments in spectral analysis of random matrices to derive accurate expressions for random projection matrices.
result Provides precise expressions that reflect the practical performance of sketching methods, including Gaussian and Rademacher sketches.
New method improves matrix completion accuracy, especially in noisy data.
problem Noisy matrix completion in recommendation systems and signal processing.
method Residual Spectral Matching criterion and pseudo-gradient algorithms.
result Improved numerical performance in noisy data environments.
Efficiently compress pretrained models using RSI for improved predictive accuracy.
problem Efficiently compressing large pretrained models for practical deployment.
method Randomized subspace iteration (RSI) for low-rank approximation of pretrained models.
result RSI achieves near-optimal approximation quality and outperforms RSVD in predictive accuracy.
New spectral methods improve matrix estimation in RL with low-rank structure.
problem Estimating matrices with low-rank structure in reinforcement learning.
method Spectral-based matrix estimation approaches.
result Spectral methods efficiently recover singular subspaces and minimize entry-wise error.
Low-rank matrix is desired in many machine learning and computer vision problems. Most of the recent studies use the nuclear norm as a convex surrogate of the rank operator. However, all singular values are simply added together by the nuclear norm, and thus the rank may not be well approximated in practical problems. …
New methods rank players using covariates and comparisons, outperforming existing algorithms.
problem Ranking players based on incomplete and noisy pairwise comparisons.
method Three spectral ranking methods incorporating player covariates.
result Proposed methods outperform existing algorithms in simulations.
Solving symmetric positive definite linear problems is a fundamental computational task in machine learning. The exact solution, famously, is cubicly expensive in the size of the matrix. To alleviate this problem, several linear-time approximations, such as spectral and inducing-point methods, have been suggested and a…
Partial convexification improves tractability of low-rank spectral optimization problems.
problem Minimizing linear objectives subject to matrix inequalities and low-rank constraints.
method Partial convexification of the domain set, deriving rank bounds, and developing a column generation algorithm.
result The partial convexification LSOP-R is equivalent to the original LSOP under certain conditions and yields high-quality solutions.
We describe several algorithms for matrix completion and matrix approximation when only some of its entries are known. The approximation constraint can be any whose approximated solution is known for the full matrix. For low rank approximations, similar algorithms appears recently in the literature under different name…
Many applications require recovering a ground truth low-rank matrix from noisy observations of the entries, which in practice is typically formulated as a weighted low-rank approximation problem and solved by non-convex optimization heuristics such as alternating minimization. In this paper, we provide provable recover…
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). Paper improves MVSC using tensor low-rank modeling.
problem Improving multi-view spectral clustering.
method Structured tensor low-rank norm for MVSC optimization.
result Proposed method outperforms state-of-the-art methods.
Study optimal algorithms for recovering signals through inhomogeneous low-rank channels.
problem Recovering signals through an inhomogeneous low-rank matrix channel.
method Derive and analyze an approximate message-passing algorithm (AMP) and a spectral method.
result The AMP iteration matches the conjectured optimal computational phase transition.
This paper improves spectral clustering for large datasets using the Nystrom method.
problem Spectral clustering's scalability issues with large datasets.
method A principled spectral clustering algorithm exploiting Nystrom approximation's spectral properties.
result Improved spectral clustering efficiency and accuracy compared to existing methods.
The paper examines how gradient descent stabilizes low-rank matrix factorization in noisy conditions.
problem Stability of low-rank implicit regularization in perturbed deep matrix factorization.
method Derives spectral conditions for gradient descent to exhibit a low-rank phase in noiseless settings and analyzes perturbed dynamics.
result Gradient descent converges to a low-rank solution under perturbation, with explicit dependence on perturbation size.
The study reveals the spectral structure of attention layers and its implications for generalization.
problem Understanding the spectral structure and generalization of trained attention layers.
method Empirical risk minimization in a single-head tied-attention layer, using random matrix theory, spin-glass theory, and approximate message passing.
result Exact high-dimensional characterization of training and test error, interpolation and recovery thresholds, and spectrum of the key and query matrices.
New algorithm learns low-rank matrices with linear number of samples.
problem Learning low-rank matrices efficiently in latent-variable applications.
method Proposed algorithm that uses linear number of samples in high dimension.
result Learning kimesk, rank-r, matrices requires $Ω( rac{kr}{ε^2})$ samples. Paper proposes a new method to improve clustering ensemble performance.
problem Improving clustering ensemble performance by refining co-association matrix.
method Low-rank tensor approximation to derive coherent-link matrix and refine co-association matrix.
result The proposed method achieves breakthrough in clustering performance compared to state-of-the-art methods.
Sketching reduces data size for accurate spectral estimation.
problem Estimating spectral density from large simulation datasets.
method Sketching for dimensionality reduction and data compression.
result Sketching provides 90% accurate spectral density estimate with 10% data.
SON-NMF estimates nonnegative rank on-the-fly for NMF.
problem Estimating the nonnegative rank of data in NMF.
method Sum-of-norms (SON) regularization to reduce rank, combined with a first-order BCD algorithm.
result SON-NMF can automatically estimate the rank from data without prior knowledge.
Spectral algorithm reduces samples needed for multitask regression.
problem Jointly recover shared and task-specific components in low-rank multitask regression.
method Common mechanism regression (CMR) model with a non-iterative spectral algorithm.
result Provable non-convex bi-linear structure is overcome with spectral algorithm.
Develops methods to estimate high rank tensors from noisy data.
problem Estimating high rank tensors from noisy observations.
method Generative latent variable tensor model, polynomial-time spectral algorithm.
result Achieves computationally optimal rate for signal tensor estimation.
A new method for nonstationary Gaussian processes using Fourier features.
problem Efficient simulation of nonstationary Gaussian processes with high-dimensional distributions.
method Discretizes the spectral representation of nonstationary processes, avoiding probability measure assumptions.
result An efficient low-rank approximation of nonstationary spectral densities, consistent and positive semi-definite.
Proposes a framework to extract ordered eigenfunctions from contextual kernels.
problem Lack of exact spectral decomposition in existing methods.
method Modular building blocks for compatibility with contextual kernels and scalability.
result Extracted eigenfunctions provide effective importance scores for feature selection.
SRF improves kernel approximation and GP regression performance.
problem Efficient kernel approximation and Bayesian kernel learning in large-scale regression problems.
method Stein variational gradient descent to generate high-quality random features and approximate spectral measure posteriors.
result SRF outperforms traditional approaches in kernel approximation and GP regression.
Paper tackles fair low-rank approximation and column subset selection.
problem Minimize loss over sub-populations in machine learning.
method Developed algorithms for fair low-rank approximation and fair column subset selection.
result Achieved polynomial time algorithms for fair low-rank approximation.
In this paper, we study the problem of approximately computing the product of two real matrices. In particular, we analyze a dimensionality-reduction-based approximation algorithm due to Sarlos [1], introducing the notion of nuclear rank as the ratio of the nuclear norm over the spectral norm. The presented bound has i…
Low-rank approximation is an effective model compression technique to not only reduce parameter storage requirements, but to also reduce computations. For convolutional neural networks (CNNs), however, well-known low-rank approximation methods, such as Tucker or CP decomposition, result in degraded model accuracy becau…
We prove, using the subspace embedding guarantee in a black box way, that one can achieve the spectral norm guarantee for approximate matrix multiplication with a dimensionality-reducing map having m=O(r~/ε2) rows. Here r~ is the maximum stable rank, i.e. squared ratio of Frobenius and op…
Paper explores statistical and computational limits of estimating low-rank Gaussian mixtures.
problem Estimating low-rank matrix-variate observations with optimal statistical and computational limits.
method Low-rank Gaussian mixture model (LrMM) and minimax lower bounds.
result Minimax optimality of maximum likelihood estimator and spectral aggregation method.
In this paper, we propose a low-rank approximation method based on discrete least-squares for the approximation of a multivariate function from random, noisy-free observations. Sparsity inducing regularization techniques are used within classical algorithms for low-rank approximation in order to exploit the possible sp…
New method reduces uncertainty in high-dimensional circuits by automatically determining tensor rank and adaptive sampling.
problem Uncertainty quantification in high-dimensional circuits due to fabrication process variations.
method Tensor regression with ℓq/ℓ2 group-sparsity regularization for rank determination and adaptive sampling. result Captures uncertainty with only 100-600 simulation samples for 19-100 random variables.
We introduce a new parameterization method for deep learning layers using spectral tensor train decomposition.
problem Efficiency and stability in deep learning models with weight matrix compression.
method Spectral Tensor Train Parameterization (STTP) of weight matrices.
result Improved compression and training stability in neural networks.
Study finds the minimum number of finite Gaussian mixtures for best approximation.
problem Finding the minimum number of finite Gaussian mixtures for best approximation.
method Local moment matching for upper bound and spectral analysis for lower bound.
result Corrects a previous lower bound in the case of Gaussian mixing distributions.
Multi-view spectral clustering, which aims at yielding an agreement or consensus data objects grouping across multi-views with their graph laplacian matrices, is a fundamental clustering problem. Among the existing methods, Low-Rank Representation (LRR) based method is quite superior in terms of its effectiveness, intu…
Matrix approximation is a common tool in machine learning for building accurate prediction models for recommendation systems, text mining, and computer vision. A prevalent assumption in constructing matrix approximations is that the partially observed matrix is of low-rank. We propose a new matrix approximation model w…
New algorithm for weighted low rank approximation with provable guarantees.
problem Weighted low rank approximation (WLRA) is computationally hard.
method Reweights the low rank solution using the weight matrix itself.
result Provably optimal approximation guarantees for WLRA.
In this paper, we study the popularly dubbed matrix completion problem, where the task is to "fill in" the unobserved entries of a matrix from a small subset of observed entries, under the assumption that the underlying matrix is of low-rank. Our contributions herein, enhance our prior work on nuclear norm regularized …