The study examines when MAML's objective has a benign landscape.
problem Understanding when MAML's objective landscape is benign.
method Analyzing the landscape of MAML objective on LQR tasks.
result The benign landscape of the MAML objective depends on task similarities.
Study optimization landscapes for overcomplete representations, showing benign geometric structures.
problem Optimizing overcomplete representations in high-dimensional data analysis.
method Formulate as ℓ 4 \ell^4 ℓ 4 -norm optimization problems with spherical constraint, analyze geometric properties. result Nonconvex objectives have benign geometric structures, ensuring local search algorithms find target solutions.
Paper analyzes Transformer learning dynamics, proving benign landscape for in-context learning.
problem Understanding how Transformers learn in context with nonlinear features.
method Mean-field and two-timescale analysis of Transformer dynamics, proving nonconvex but benign landscape.
result Proves mean-field dynamics avoid saddle points, leading to improved optimization.
Wide deep neural networks are easy to optimize without constraints.
problem Optimizing wide deep neural networks.
method Analysis of optimization landscapes and empirical-risk minimization.
result Wide neural networks have no confined points, making optimization easier.
The paper analyzes phase retrieval under limited samples, ensuring a benign local landscape for convergence.
problem Ensuring a benign local landscape for phase retrieval under limited samples.
method Fine-grained analysis of local landscape properties under the regime of limited samples.
result Gradient descent can converge to an o d ( 1 ) o_d(1) o d ( 1 ) -loss solution exponentially fast under certain conditions. New findings on optimization landscape of Toeplitz covariance estimation.
problem Understanding the geometry of the Gaussian maximum-likelihood objective for Toeplitz covariance estimation.
method Overparameterized Carathéodory representation of positive definite Toeplitz covariance matrices, focusing on both amplitudes and frequencies.
result Joint optimization of amplitudes and frequencies leads to a benign population landscape, allowing for global recovery of the true Toeplitz covariance.
This paper introduces SRPR for robust phase retrieval with smoothed loss functions.
problem Robust phase retrieval from noisy quadratic measurements with corruptions.
method Smoothed robust phase retrieval (SRPR) using convolution-type smoothed loss functions.
result SRPR has no spurious local solutions and benign landscape under corruptions.
Deeper models have a more favorable optimization landscape, making them more robust to noise.
problem Characterizing the effect of depth on the optimization landscape of linear regression models.
method Robust and over-parameterized setting, simple sub-gradient method.
result A simple sub-gradient method converges to a balanced solution that is close to the ground truth and enjoys a flat local landscape.
Optimizes MMD learning for generative models with theoretical guarantees.
problem Theoretical guarantees for optimizing non-convex MMD objectives.
method Analyzes MMD optimization landscape for specific distributions.
result Gradient-based methods globally minimize MMD objective for certain distributions.
This work analyzes a two-stage algorithm for single index models, showing precise asymptotics of gradient descent.
problem Learning single index models with non-convex optimization.
method Spectral initialization followed by gradient descent, with detailed analysis of dynamics and asymptotics.
result Gradient descent converges to long-time fixed points in the large system limit, representing mean field behavior.
New theory shows predictive coding makes learning landscape easier to navigate.
problem Understanding the impact of predictive coding's inference procedure on learning efficiency.
method Analyzed the geometry of the energy landscape of deep linear networks, proving many non-strict saddles become strict in the equilibrated energy.
result All highly degenerate (non-strict) saddles of the loss become strict in the equilibrated energy, suggesting a more robust learning landscape.
WSFN overcomes saddle points for non-convex functionals in Wasserstein space.
problem Minimizing non-convex functionals over the Wasserstein space with saddle point avoidance.
method WSFN is a second-order method that preconditions the Wasserstein gradient to avoid saddle points.
result WSFN escapes saddle regions and reaches a global minimizer in polynomial time.
FPCA optimizes fairness in target vectors' span.
problem Fairness in principal component analysis for multiple target vectors.
method Non-concave maximization of worst projected target norm using sub-gradient descent.
result Optimization landscape is benign with globally optimal local minima.
Improved optimization guarantees for deep learning models with Nesterov acceleration.
problem Optimization in non-convex deep learning landscapes.
method Analysis of Nesterov acceleration in benignly non-convex landscapes.
result Identical guarantees can be obtained in optimization problems with weak geometric assumptions, especially in overparametrized deep learning.
We examine the squared error loss landscape of shallow linear neural networks. We show---with significantly milder assumptions than previous works---that the corresponding optimization problems have benign geometric properties: there are no spurious local minima and the Hessian at every saddle point has at least one ne…
Neural collapse occurs in normalized features over a Riemannian manifold.
problem Understanding neural collapse in normalized feature models.
method Simplified multi-class classification task to a nonconvex optimization problem over the Riemannian manifold, analyzing the landscape of critical points.
result The only global minimizers are neural collapse solutions, with all other critical points being strict saddles.
Algorithm aligns 3D density maps using Wasserstein distance.
problem Aligning 3D density maps in cryogenic electron microscopy.
method Minimizing 1-Wasserstein distance after rigid transformation using Bayesian optimization.
result Improved accuracy and efficiency in protein molecule alignment.
Sharp condition found for Burer-Monteiro method to work for MaxCut-type SDPs.
problem MaxCut-type semidefinite programs with low-rank solutions.
method Sharp condition on Laplacian matrix conditioning for global minimizers of non-convex problem.
result Any second-order critical point is a global minimizer under the given condition.
Study on benign overfitting in leaky ReLUs with moderate input dimensions.
problem Understanding when overfitting is beneficial in neural networks.
method Two-layer leaky ReLU networks trained with hinge loss, considering signal-to-noise ratio.
result Characterization of conditions for benign overfitting based on signal-to-noise ratio.
BOinG optimizes HPO problems by focusing on promising local regions.
problem Expensive black-box optimization problems, especially hyperparameter optimization.
method Two-stage approach: global surrogate model followed by local model.
result BOinG exploits the structure of typical HPO problems and performs well on mid-sized problems.
We provide new theoretical insights on why over-parametrization is effective in learning neural networks. For a k k k hidden node shallow network with quadratic activation and n n n training data points, we show as long as k ≥ 2 n k \ge \sqrt{2n} k ≥ 2 n , over-parametrization enables local search algorithms to find a \emph{globally} op…
Recent results in the literature indicate that a residual network (ResNet) composed of a single residual block outperforms linear predictors, in the sense that all local minima in its optimization landscape are at least as good as the best linear predictor. However, these results are limited to a single residual block …
New framework analyzes effectiveness of neural network-based combinatorial problem solvers.
problem Analyzing neural network-based methods for combinatorial optimization problems.
method Introducing a theoretical framework to assess the effectiveness of solution-samplers using policy-gradient methods.
result Positive theoretical answer to the existence of expressive, tractable, and benign optimization landscapes for combinatorial problems.
Attention models can overfit without harming test performance.
problem Understanding benign overfitting in single-head attention models.
method Analyzing conditions for benign overfitting in a single-head softmax attention model.
result A single-head attention model can overfit without harming test performance under certain conditions.
ResNet models overfit benignly on Cifar10 but not on ImageNet due to label noise.
problem Understanding why benign overfitting fails in real-world classification tasks with label noise.
method Theoretical analysis of benign overfitting under a mild overparameterization setup.
result Benign overfitting can fail in the presence of label noise, unlike in heavy overparameterization settings.
Paper finds conditions for benign overfitting in neural networks.
problem Benign overfitting in leaky ReLU two-layer neural networks.
method Established directional convergence and classification error bounds.
result Benign overfitting occurs with high probability on mixture data.
New insights into when benign overfitting occurs in linear and classification tasks.
problem Understanding when benign overfitting happens in linear and classification models.
method Analysis of a generic data model and comparison of predictors (minimum-norm interpolating and max-margin).
result The minimum-norm interpolating predictor is biased towards an inconsistent solution, preventing benign overfitting in linear regression.
New insights into why neural networks can overfit without interpolating data.
problem Understanding why neural networks can overfit without interpolating data in fixed dimensions.
method Analyzing the smoothness of estimators and their derivatives.
result Benign overfitting is possible with estimators that have large enough derivatives, not just in high dimensions but also in fixed dimensions.
Study on size and depth of neural networks for approximating benign functions, showing barriers and explicit results.
problem Understanding how size and depth of neural networks affect their ability to approximate benign functions.
method Analyzing ReLU networks for benign functions, proving barriers and explicit results.
result Explicit benign functions that cannot be approximated by networks of certain sizes or depths, showing barriers to size and depth separation.
Two-layer CNNs can overfit without harm under certain conditions.
problem Understanding when and how overfitting occurs in neural networks.
method Theoretical analysis of a two-layer CNN trained by gradient descent.
result A sharp phase transition between benign and harmful overfitting based on signal-to-noise ratio.
We analyze neural collapse in neural networks, showing that features collapse to vertices of a Simplex ETF.
problem Understanding and optimizing the features learned in the last layer of neural networks during training.
method Simplified unconstrained feature model, studying the global optimization landscape of cross-entropy loss with weight decay.
result The global minimizers of the loss are Simplex ETFs, and other critical points are strict saddles with negative curvature.
New framework solves dynamic bilevel optimization problems in reinforcement learning.
problem Dynamic objective functions in reinforcement learning and human feedback.
method Principled penalty-based methods for bilevel reinforcement learning.
result Demonstrated effectiveness of penalty-based algorithms in simulations.
Two-layer ReLU networks can overfit without harm, study finds.
problem Understanding when and how two-layer ReLU networks can overfit without harming generalization.
method Established algorithm-dependent risk bounds for two-layer ReLU convolutional neural networks with label-flipping noise.
result Gradient descent-trained ReLU networks can achieve near-zero training loss and Bayes optimal test risk.
Meta learning works well with overparameterized models, a phenomenon called 'benign overfitting'.
problem Understanding why overparameterized models perform well in few-shot learning.
method Analyzed the generalization performance of gradient-based meta learning with an overparameterized meta linear regression model.
result Demonstrated that overparameterized meta learning can still generalize well, a phenomenon called 'benign overfitting'.
Deep neural networks can generalize well even with perfect fits to noisy data.
problem Understanding the conditions under which deep neural networks generalize well in the presence of noise.
method Comprehensive study of linear maximum margin classifiers, focusing on noisy and noiseless cases.
result Discovery of a phase transition in test error bounds for the noisy model.
The paper explores how benign overfitting occurs in heavy-tailed input distributions.
problem Understanding overfitting in heavy-tailed input distributions.
method Analysis of maximum margin classifiers on unregularized logistic loss with gradient descent.
result Linear classifiers trained under certain conditions can asymptotically achieve the noise level as misclassification error.
New analysis shows bias term affects conditions for benign overfitting in linear classifiers.
problem Understanding conditions for good generalization in linear classifiers with bias terms.
method Extending Hashimoto et al.'s results to include bias terms, analyzing covariance structure and label noise.
result Benign overfitting persists in linear classifiers with bias terms, with new constraints on data's covariance structure.
The paper explores how noise in features can lead to benign overfitting in machine learning models.
problem Understanding the conditions for benign overfitting in machine learning models.
method Examined random feature models, specifically two-layer neural networks with fixed first layer weights, and analyzed the role of noise in features.
result Noise in features plays an important implicit regularization role in the phenomenon of benign overfitting.
ELS framework improves safety alignment by dynamically steering LLMs towards helpful responses.
problem Over-Refusal in Aligned Large Language Models
method Fine-tuning free framework using an Energy-Based Model (EBM) to dynamically steer LLMs during inference.
result Extensive experiments show a significant reduction in false refusals (from 57.3% to 82.6%) while maintaining safety performance.
Adversarial training can lead to overfitting without compromising robustness.
problem Explaining benign overfitting in adversarially robust linear classification.
method Theoretical analysis and numerical experiments on adversarial training.
result Adversarially trained linear classifiers can achieve near-optimal risks despite overfitting noisy data.
Neural networks can interpolate noisy data and still generalize well.
problem Generalization of neural networks trained on noisy data.
method Two-layer neural networks trained to interpolation by gradient descent on corrupted labels.
result Neural networks can achieve zero training error and optimal test error.
New neural networks learn single-index models efficiently.
problem Learning low-dimensional structure in high-dimensional data.
method Shallow neural networks with frozen biases, studied via gradient flow.
result Generalization guarantees match near-optimal sample complexity.
Unified framework explains why overfitting is benign in interpolating learning.
problem Understanding why overfitting is benign in highly overparameterized models.
method Spectral-transport stability framework.
result Sharp benign-overfitting criterion and explicit phase-transition rates.
Nonnegative low-rank matrix recovery can have spurious local minima.
problem Nonnegative low-rank matrix recovery problems can have spurious local minima.
method Investigated projected gradient methods for nonnegative low-rank recovery problems.
result Benign nonconvexity holds in the fully-observed case with RIP constant δ=0 but fails in the partially-observed case and higher-rank ground truths.
Study reveals learning curves and benign overfitting in spectral algorithms for large dimensions.
problem Understanding learning curves and benign overfitting in spectral algorithms for large-dimensional data.
method Analysis of learning curves and benign overfitting in spectral algorithms for inner-product kernels on the sphere and general domains.
result Characterization of three distinct regimes: over-regularized, under-regularized, and interpolation regimes, revealing benign overfitting across both under-regularized and interpolation regimes.
Linear models can overfit without harming OOD generalization under certain conditions.
problem Understanding how overparameterized linear models generalize to out-of-distribution data.
method Analyzing overparameterized linear models under covariate shift, providing guarantees for OOD generalization.
result Benign overfitting occurs in standard ridge regression under OOD conditions, with specific structural conditions on target covariance.
New study finds many neural networks are not benignly overfitting.
problem Understanding the behavior of overfitting in neural networks.
method Exploring kernel ridge regression and deep neural networks to identify overfitting behaviors.
result Many interpolating methods, including neural networks, exhibit tempered overfitting rather than benign or catastrophic.
ABO extends RLS for online learning in non-stationary time-series, improving accuracy and speed.
problem Online learning in non-stationary time-series with overparameterized models.
method QR-based exponentially weighted RLS algorithm with orthogonal-triangular updates.
result ABO maintains bounded residuals and stable condition numbers while achieving speed improvements.