Kernel ridgeless regression with random features shows good generalization without explicit regularization.
problem Generalization of kernel ridgeless regression without explicit regularization.
method Investigation of ridgeless regression with random features and stochastic gradient descent, exploring the effect of random features error and spectral density optimization.
result Random features error exhibits the double-descent curve, leading to improved generalization.
Study shows that ridgeless Gaussian kernel regression overfits even with varying bandwidth or dimensionality.
problem Analyzing overfitting in Gaussian kernel ridgeless regression with varying bandwidth or dimensionality.
method Examined the behavior of minimum norm interpolating solutions for fixed and increasing dimensions under varying bandwidth and sample size.
result Ridgeless solutions are never consistent and can be worse than null predictor with large enough noise, even with varying bandwidth or dimensionality.
Paper explores how DPP sampling can implicitly regularize kernel regression.
problem Improving kernel regression by reducing redundancy in data.
method Using Determinantal Point Processes (DPPs) to sample subsets implicitly regularizes ridgeless Kernel Regression.
result Ensemble of ridgeless regressors can be effective for datasets with redundant information.
Enhanced kernel ridgeless regression improves performance with LAB RBF kernels.
problem Lack of flexibility in kernel ridgeless regression.
method Locally-Adaptive-Bandwidths (LAB) RBF kernels and kernel learning techniques.
result Functions learned from LAB RBF kernels belong to an integral space of RKHSs, demonstrating robust generalization.
Lower bound proves ridgeless regression performs poorly near interpolation threshold.
problem Proving performance of ridgeless regression near interpolation threshold.
method Distribution-independent lower bound for mean squared error in noisy ridgeless linear regression.
result Lower bound implies ridgeless regression performs poorly near interpolation threshold.
The paper examines prediction and estimation risks of ridgeless least squares under general error assumptions.
problem Prediction and estimation risks of ridgeless least squares under realistic error structures.
method Analysis of prediction and estimation risks under general regression error assumptions, including clustered or serial dependence.
result The benefits of overparameterization extend to time series, panel, and grouped data.
Downsampling can improve generalization in ridgeless linear regression, especially with optimal sketching size.
problem Improving generalization in ridgeless linear regression with limited data.
method Investigating the effects of downsampling on the sketched ridgeless least square estimator in the proportional regime.
result Optimal sketching size minimizes out-of-sample prediction risks and stabilizes risk curves.
Data splitting enhances model performance in overparametrized ridgeless regression.
problem Computational inefficiency in training models with large datasets.
method Data splitting as a regularization technique in overparametrized ridgeless regression.
result Data splitting improves statistical performance and computational complexity.
Ridgeless ReLU networks interpolate datasets and extrapolate based on curvature signs.
problem Interpolating and extrapolating 1D datasets with ReLU networks.
method Minimizes ℓ2-norm of weights, extrapolates based on curvature signs. result Ridgeless ReLU interpolants extrapolate as nearest neighbor curvature extrapolation.
New bounds for KRR condition number reveal overfitting phenomena.
problem Characterizing overfitting in KRR with varying kernel spectral decay.
method Derived new bounds for kernel matrices, enhanced test error bounds, and identified feature independence role.
result Identified tempered and catastrophic overfitting phenomena.
Interpolating models can have heavy-tailed risk, leading to rare but severe errors.
problem Interpolating models' tail risk is poorly understood, affecting rare but impactful errors.
method Large-deviation methods to study the fragility of high-dimensional linear interpolators.
result Ridgeless regression exhibits heavy-tailed risk, while ridge-regularized estimators have better tail behavior.
In the absence of explicit regularization, Kernel "Ridgeless" Regression with nonlinear kernels has the potential to fit the training data perfectly. It has been observed empirically, however, that such interpolated solutions can still generalize well on test data. We isolate a phenomenon of implicit regularization for…
The paper analyzes bagging in overparameterized learning, deriving risk properties and optimal subsample sizes.
problem Characterizing the risk of bagged predictors in overparameterized settings.
method General strategy using classical results on simple random sampling, specialized for ridge and ridgeless predictors.
result Derives exact asymptotic risk of bagged ridge and ridgeless predictors under various conditions.
Sharp analysis of knowledge distillation for high-dimensional regression.
problem Characterizing the risk of target models in high-dimensional settings.
method Sharp non-asymptotic bounds for ridgeless regression under model and distribution shifts.
result Identifies optimal surrogate models and reveals benefits and limitations of discarding weak features.
Study ridge ensembles in proportional feature-to-sample size regime, proving risk equivalence and GCV consistency.
problem Characterizing and optimizing ridge ensembles in proportional feature-to-sample size regimes.
method Proportional asymptotics analysis, GCV for tuning, proving risk equivalence.
result Risk of optimal full ridgeless ensemble matches optimal ridge predictor's risk.
Knoop enhances variable selection with over-parameterization and knockoffs.
problem Challenges of variable selection in high-dimensional datasets.
method Generates knockoff variables, integrates them into an over-parameterized model, and uses anomaly-based significance tests.
result Superior performance in variable selection compared to existing methods.
Interpolators -- estimators that achieve zero training error -- have attracted growing attention in machine learning, mainly because state-of-the art neural networks appear to be models of this type. In this paper, we study minimum ℓ2 norm ("ridgeless") interpolation in high-dimensional least squares regression. …
New method estimates precision matrices without models, achieving dense, consistent, and model-free properties.
problem Lack of methods that are dense, consistent, and model-free for precision matrix estimation.
method General class of estimators that unify dense, consistent, and model-free properties within a nonasymptotic framework.
result Ridgeless regression exhibits the double descent phenomenon, establishing a precision matrix analogue to linear regression's double descent.
Locality helps in learning from high-dimensional data.
problem Understanding how convolutional neural networks learn from high-dimensional data.
method Teacher-student framework for kernel regression with convolutional kernels.
result Locality is key to determining the learning curve exponent in high-dimensional data.
Optimal recovery framework for non-IID data in Hilbert spaces.
problem Generalization in non-IID data scenarios.
method Optimal recovery perspective, semidefinite programming, kernel ridgeless regression.
result Optimal recovery formula coincides with kernel ridgeless regression in some cases.
Overparameterized ensembles don't offer generalization benefits over single large models.
problem Theoretical limitations of ensembles in overparameterized settings.
method Using ensembles of random feature (RF) regressors, the paper clarifies how modern ensembles differ from underparameterized counterparts.
result Infinite ensembles of overparameterized RF regressors become pointwise equivalent to single infinite-width RF regressors, and finite width ensembles converge to single models with the same parameter budget.
Lecture notes on advanced linear regression methods.
problem Understanding the properties of linear regression estimators in high dimensions.
method Proposition-proof exploration of least squares, ridgeless, ridge, and lasso estimators.
result Detailed analysis of the existence, uniqueness, relations, computation, and non-asymptotic properties of these estimators.
W2S FT often outperforms weak teachers due to low intrinsic dimensionality.
problem Understanding why weak-to-strong finetuning outperforms weak models.
method Analyzing W2S in ridgeless regression setting, focusing on variance reduction.
result Weak teacher's variance is inherited by strong student in shared feature subspace, reduced in discrepancy subspace.
This work analyzes how preconditioning affects generalization in machine learning models.
problem The impact of preconditioning on the generalization of machine learning models.
method An asymptotic bias-variance decomposition of the generalization error for ridgeless regression under various preconditioners.
result The optimal preconditioner depends on label noise, model specification, and signal alignment, with NGD potentially better under certain conditions.
The paper analyzes how generated data improves adversarial training in high-dimensional regression.
problem Improving adversarial training in high-dimensional regression.
method Theoretical analysis of a two-stage training approach with generated data and pseudo-labels.
result Two-stage adversarial training achieves better performance than ridgeless training in high-dimensional linear regression.
Study the cost of overfitting in noisy KRR models.
problem Cost of overfitting in noisy kernel ridge regression.
method An agnostic view of overfitting cost as a function of sample size for any target function, using Gaussian universality ansatz and task eigenstructure.
result Characterization of benign, tempered, and catastrophic overfitting.
Complexity helps identify sparse risk factors in asset pricing.
problem Tension between feature richness and economic parsimony in high-dimensional asset pricing.
method Expanding feature space and using basis pursuit to discover sparse risk factors.
result Nonlinear feature expansions combined with basis pursuit yield superior out-of-sample performance.
Linear models can be poisoned by shifting a fraction of one class's data, revealing scaling laws and weight alignment.
problem Understanding and quantifying data poisoning in linear models.
method Analysis of ridge least squares with an unpenalized intercept, using resolvent techniques and random matrix theory.
result Closed-form limits for the poisoned score, revealing scaling laws and weight alignment with the poisoning direction.
New method stabilizes machine learning for physics-informed inverse problems.
problem Reconstructing physical quantities from PDE-compliant measurements.
method Physics-informed learning with smooth inductive bias.
result PDE operators stabilize variance and prevent overfitting in fixed dimensions.
Paper analyzes SGD in kernel regression, showing it outperforms offline methods.
problem Performance of SGD in kernel regression compared to offline methods.
method Analyzes Stochastic Gradient Descent (SGD) in kernel regression under misspecified models.
result SGD achieves min-max optimal rates up to constants, avoiding saturation.
Consider the problem: given the data pair (x,y) drawn from a population with f∗(x)=E[y∣x=x], specify a neural network model and run gradient flow on the weights over time until reaching any stationarity. How does ft, the function computed by the neural network…
We analyze optimal weighted ridge regression in overparameterized linear models.
problem Optimal regularization in overparameterized linear regression models.
method Generalized ridge regression with weighted regularization.
result The optimal regularization parameter can be negative in overparameterized settings.
Linear memory stores associations up to a logarithmic scale, but listwise retrieval can handle a quadratic scale.
problem How many key-value associations can a linear memory store?
method Analyzed linear memory models for top-1 and listwise retrieval, proving phase transitions and developing asymptotic theories.
result Linear memory has a logarithmic capacity for top-1 retrieval and a quadratic capacity for listwise retrieval.