The paper analyzes how low-rank layers in neural networks improve generalization.
problem Understanding how low-rank layers affect generalization in neural networks.
method Applying Maurer's chain rule for Gaussian complexity to analyze rank and spectral norm constraints.
result Deep networks with low-rank layers achieve better generalization than those with full-rank layers.
Low-rank matrices explain data science patterns.
problem Why do data matrices often have low rank?
method A generative model with latent variables and piecewise functions.
result Approximating large matrices with low rank is feasible.
Weight Decay induces low-rank weight matrices in neural networks, improving generalization.
problem Improving generalization in neural networks.
method Training ReLU NN with Weight Decay and Stochastic Gradient Descent.
result The weight matrix of a trained NN is approximately rank-two.
Unified framework for nonconvex low-rank matrix estimation using gradient descent.
problem Estimating low-rank matrices in noisy and noiseless settings.
method Gradient descent algorithm applied to nonconvex optimization.
result Unified framework guarantees linear convergence to the unknown low-rank matrix with optimal statistical error.
Low-rank modeling generally refers to a class of methods that solve problems by representing variables of interest as low-rank matrices. It has achieved great success in various fields including computer vision, data mining, signal processing and bioinformatics. Recently, much progress has been made in theories, algori…
New method uses low-rank tensor factor analysis for better image restoration.
problem Restoring images from limited data.
method Low-rank tensor factor analysis combined with ADMM.
result The method outperforms traditional approaches, especially at low sampling rates.
New framework for low-rank tensor analysis on graphs.
problem Low-rank tensor analysis on non-Euclidean domains.
method Graph-based low-rank decomposition and convex optimization.
result Significant speed-up and performance enhancement at low SNR.
Novel method for efficient low-rank matrix estimation and bandit algorithms.
problem Low-rank matrix estimation and bandit problems.
method LowPopArt method for low-rank matrix estimation and novel experimental design criterion.
result Improved recovery guarantees and regret bounds for low-rank bandit algorithms.
A new method for efficient neural network fine-tuning using queryable low-rank update atoms.
problem Rigidity of static low-rank adaptation methods when input and depth-wise computation vary.
method A shared queryable memory of low-rank update atoms, allowing dynamic and context-sensitive adaptation.
result Improves final test performance and training stability compared to standard low-rank adaptation.
New algorithm speeds up recovery of low-rank matrices.
problem Nonconvex low-rank matrix recovery problems.
method Stochastic variance-reduced gradient descent with semi-stochastic gradient.
result Linear convergence rate to unknown low-rank matrix.
Unified framework for solving low-rank plus sparse matrix recovery problems.
problem Solving general low-rank plus sparse matrix recovery problems.
method Unified framework based on matrix factorization, projected gradient descent, and double thresholding operator.
result Our algorithm converges to the unknown low-rank and sparse matrices at a locally linear rate, matching robustness guarantees.
The paper establishes theoretical foundations for low-rank knowledge distillation in LLMs.
problem Understanding the theoretical underpinnings of low-rank knowledge distillation in LLMs.
method Theoretical framework for low-rank knowledge distillation, including convergence rates and generalization bounds.
result Theoretical analysis reveals optimal rank r ∗ = O ( n ) r^* = O(\sqrt{n}) r ∗ = O ( n ) for minimizing generalization error. A hierarchical Gaussian prior model improves low-rank matrix completion.
problem Low-rank matrix completion with improved structure exploitation.
method Hierarchical Gaussian prior model with GAMP embedded variational Bayesian inference.
result The proposed method outperforms state-of-the-art matrix completion methods.
TensorGuide improves LoRA efficiency and expressivity through joint tensor-train optimization.
problem Limited expressivity and generalization of standard LoRA.
method TensorGuide uses a unified tensor-train structure with controlled Gaussian noise to generate correlated low-rank matrices.
result TensorGuide achieves superior accuracy and scalability with fewer parameters compared to standard LoRA and TT-LoRA.
Solves low-rank approximation problems in Hilbert spaces.
problem Low-rank approximation in Hilbert spaces.
method Closed-form solutions and error bounds for bounded linear operators.
result Generalization to bounded linear operators from finite dimensions.
Paper develops methods for non-quadratic loss low-rank matrix recovery.
problem Recovery of low-rank matrices with non-quadratic losses.
method Projected gradient method with a regularity projection oracle.
result Projected gradient method converges globally and linearly.
A new model for data with zeros or missing values.
problem Data with excess zeros or missing values.
method Composite loss framework for low-rank modeling, combining generalized low-rank and hurdle methods.
result Demonstrated on a manufacturing data set and applied to missing value imputation.
FedLoRU improves FL efficiency by using low-rank updates.
problem Communication inefficiency and performance reduction in Federated Learning.
method Proposes FedLoRU, a low-rank update framework for FL, which reduces communication costs while maintaining performance.
result FedLoRU achieves convergence rates similar to FedAvg and is robust to heterogeneous and large numbers of clients.
Paper studies asymmetric matrix sensing, proving gradient descent converges to low-rank solutions.
problem Reconstructing asymmetric low-rank matrices from linear measurements.
method Factorized gradient descent with coupling and regularization properties.
result Gradient descent from small random initialization converges to globally optimal and generalizing solutions.
New algorithm tackles low-rank constraints in optimal transport problems.
problem Optimal transport problems with low-rank constraints.
method Explicit factorization of low-rank couplings as a product of sub-coupling factors linked by a common marginal.
result Stationary convergence of the algorithm proved.
Distributed algorithm finds global solutions for low-rank matrices.
problem Finding global solutions for low-rank matrices in distributed systems.
method Distributed Gradient Descent (DGD+) with LOCAL variables.
result DGD+LOCAL converges to global minimizer with exact consensus.
New nonconvex regularizer speeds up low-rank matrix completion.
problem Low-rank matrix completion with good theoretical and empirical performance.
method Proposes a new nonconvex regularizer with adaptive shrinkage, scalable, and fast optimization.
result Proposed method achieves state-of-the-art recovery performance and is the fastest.
Bayesian model improves image completion accuracy by automatically learning low rank structure.
problem Improving image completion accuracy with limited data and avoiding overfitting.
method Developed a Bayesian low rank tensor ring model with multiplicative interaction and Student-T distribution for sparse core factors.
result The proposed method outperforms state-of-the-art image completion techniques, especially in recovery accuracy.
FLAMBE tackles RL in low rank MDPs by learning features.
problem Dealing with the curse of dimensionality in RL.
method Develops FLAMBE, a method that engages in exploration and representation learning for RL in low rank transition models.
result FLAMBE efficiently learns features for RL in low rank transition models.
New method solves nonsmooth low-rank matrix optimization problems efficiently.
problem Nonsmooth and low-rank matrix optimization problems in statistics and machine learning.
method Low-rank Extragradient Method with warm-start initialization.
result The extragradient method converges to an optimal solution with rate O ( 1 / t ) O(1/t) O ( 1/ t ) and requires only two low-rank SVDs per iteration. New approach to convex hulls for low-rank problems.
problem Characterizing convex hulls for low-rank sets.
method Matrix perspective function and orthogonal projection matrices.
result Strong relaxations for various low-rank problems.
A new method for matrix completion identifies low-rank submatrices.
problem Matrix completion for non-low-rank matrices.
method Targeted framework: extract low-rank submatrices, complete separately.
result Significantly smaller reconstruction errors than classical methods.
Consider a movie recommendation system where apart from the ratings information, side information such as user's age or movie's genre is also available. Unlike standard matrix completion, in this setting one should be able to predict inductively on new users/movies. In this paper, we study the problem of inductive matr…
Unified approach tackles high-dimensional tensor bandits with convex optimization and weakly decomposable regularizers.
problem Challenges in high-dimensional generalized tensor bandits where existing algorithms fail.
method Proposes a generalized linear tensor bandits algorithm with a unified analytical framework using convex optimization and weakly decomposable regularizers.
result Unified analytical framework provides better bounds and broader applicability compared to existing methods.
Language models exhibit low-rank structure, which can be used for generation.
problem Understanding the low-dimensional structure of large language models.
method Empirical demonstration and theoretical analysis of the approximate rank of language models' logits.
result Language models can generate responses using linear combinations of unrelated prompts.
CoreFlow models matrix-valued distributions efficiently, preserving shared low-rank structure.
problem Learning matrix-valued distributions from high-dimensional and incomplete data.
method Low-rank flow model that learns shared row/column subspaces and trains a normalizing flow on the core.
result CoreFlow improves generation quality in few-sample regimes and remains competitive in data-rich settings.
New framework explains why nonconvex methods work well in low-rank matrix estimation.
problem Nonconvex low-rank matrix estimation problems in machine learning.
method Developed a theoretical framework revealing a benign regularizer.
result Nonconvex procedures can behave well due to a disguised convexity.
Proposes GTTN for discovering all low-rank structures in deep multi-task learning.
problem Discovering all low-rank structures among tasks in deep multi-task models.
method Introduces GTTN, a convex combination of matrix trace norms of all tensor flattenings, to automatically determine the importance of components.
result Demonstrates the effectiveness of GTTN on real-world datasets.
Paper introduces G-LowTESTR for efficient tensor bandits.
problem Efficient decision-making in multi-dimensional data with non-linear reward functions.
method Generalized low-rank tensor contextual bandits model and G-LowTESTR algorithm.
result G-LowTESTR achieves superior regret bound compared to vectorization and matricization methods.
Paper develops a new weighted low-rank matrix approximation technique.
problem Matrix completion with missing data.
method Element-wise weighted generalization of low-rank matrix approximation.
result Proposes an algorithm and acceleration techniques for solving the weighted problem.
Equivalent formulations for low-rank matrix optimization are proven.
problem Low-rank matrix optimization with rank constraints.
method Established geometric landscape connections between manifold and factorization formulations.
result Equivalence between manifold and factorization formulations at FOSPs, SOSPs, and strict saddles.
This work presents a general framework for solving the low rank and/or sparse matrix minimization problems, which may involve multiple non-smooth terms. The Iteratively Reweighted Least Squares (IRLS) method is a fast solver, which smooths the objective function and minimizes it by alternately updating the variables an…
Principal components analysis (PCA) is a well-known technique for approximating a tabular data set by a low rank matrix. Here, we extend the idea of PCA to handle arbitrary data sets consisting of numerical, Boolean, categorical, ordinal, and other data types. This framework encompasses many well known techniques in da…
We introduce a new framework for optimal transport using Schatten-p regularization to recover low-rank structures.
problem Optimal transport problems with low-rank structure recovery.
method Schatten-p norm regularization to promote low-rank structure in transport maps and plans.
result Unified convex programs for low-rank structure recovery with theoretical guarantees and efficient algorithms.
Paper proposes a method for estimating complex low-rank matrices from phase-only measurements.
problem Estimating complex low-rank matrices from magnitude-only measurements.
method A hierarchical prior model with a Gaussian-Wishart distribution is used to promote low-rankness. A variational EM algorithm is developed to solve the problem.
result The proposed method is less sensitive to initialization and performs well with random initialization.
ScaledGD improves gradient descent for ill-conditioned low-rank matrix estimation.
problem Efficiently solving ill-conditioned low-rank matrix estimation problems.
method Scaled Gradient Descent (ScaledGD) with adaptive pre-conditioners.
result Linear convergence rate independent of condition number, low per-iteration cost.
TFB simplifies Bayesian LLM uncertainty estimation without extra training.
problem Estimating uncertainty in LLM responses remains challenging.
method Training-Free Bayesianization (TFB) that transforms low-rank adapters into Bayesian ones without additional training.
result TFB achieves superior uncertainty estimation and generalization compared to existing methods.
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.
An algorithm tackles low-rank linear bandit problems with improved regret bounds.
problem Low-rank linear bandit problems where rewards are inner products with an unknown low-rank matrix.
method Combines online-to-confidence-set conversion and exponentially weighted average forecaster with a covering of low-rank matrices.
result Achieves O ~ ( ( d 1 + d 2 ) 3 / 2 r T ) \widetilde{O}((d_1+d_2)^{3/2}\sqrt{rT}) O (( d 1 + d 2 ) 3/2 r T ) regret, improving over standard bounds when r ≪ min { d 1 , d 2 } r \ll \min\{d_1,d_2\} r ≪ min { d 1 , d 2 } . New model-free algorithms learn representations for low-rank MDPs efficiently.
problem Learning representations in reinforcement learning for low-rank MDPs.
method Developed minimax representation learning objective and interleaved with reward-free exploration.
result Proven sample efficiency and scalability to complex environments.
Paper tackles matrix completion with a mixture of low-rank matrices.
problem Matrix completion with a mixture of low-rank matrices.
method Generalized matrix completion model (MMC) with theoretical and practical contributions.
result MMC provides a more accurate model for recommender systems and clustering.
ScaledGD accelerates ill-conditioned low-rank estimation.
problem Slow convergence of gradient descent in ill-conditioned problems.
method Scaled gradient descent (ScaledGD) with preconditioning.
result Linear convergence rate independent of condition number.
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.