L2O-CFGD meta-learns hyperparameters for FGD, improving performance.
problem Challenges in convergence and hyperparameter selection for FGD.
method Learning to Optimize Caputo Fractional Gradient Descent (L2O-CFGD).
result Meta-learned schedule outperforms static hyperparameters and achieves comparable performance to black-box meta-learners.
Paper introduces a new optimization method for imbalanced datasets.
problem Overfitting in imbalanced datasets, especially in financial fraud detection.
method Fractional Weyl Integral optimization algorithm.
result Significantly improved performance in financial fraud detection (40% improvement in PR-AUC).
Paper addresses fault-tolerance in distributed machine learning with stochastic gradient descent.
problem Fault-tolerance in distributed stochastic gradient descent (D-SGD) for machine learning.
method Proposes norm-based comparative gradient elimination (CGE) to robustify D-SGD against Byzantine faulty agents.
result CGE guarantees fault-tolerance against a bounded fraction of Byzantine agents under standard stochastic assumptions.
Single gradient step finds adversarial examples in random neural networks.
problem Finding adversarial examples in neural networks with random architectures.
method Gradient descent approach applied to random undercomplete and overcomplete two-layers neural networks.
result A single gradient step is sufficient to find adversarial examples in random neural networks.
Gradient descent amplifies random features in neural networks to useful ones.
problem Generalization in neural networks trained on corrupted data.
method Characterization of feature-learning process in two-layer ReLU networks trained by gradient descent.
result Gradient descent amplifies random features to useful ones, achieving near optimal generalization error.
Gradient descent solves robust mean estimation in high dimensions.
problem High-dimensional robust mean estimation in the presence of adversarial outliers.
method Gradient descent with a structural lemma showing near-optimal solutions.
result Gradient descent can solve the robust mean estimation problem directly.
This paper studies the problem of distributed stochastic optimization in an adversarial setting where, out of the m machines which allegedly compute stochastic gradients every iteration, an α-fraction are Byzantine, and can behave arbitrarily and adversarially. Our main result is a variant of stochastic gradient de…
New RDP guarantees for heavy-tailed SDEs and SGD.
problem Characterizing differential privacy for heavy-tailed noise in learning algorithms.
method Rényi flow computations and fractional Poincaré inequalities.
result First RDP guarantees for heavy-tailed SDEs with weaker dependence on dimension.
Gradient descent training of neural networks leads to solutions close to natural cubic splines.
problem Understanding the implicit bias of gradient descent in neural networks.
method Analysis of gradient descent training for wide neural networks, focusing on the curvature penalty and initialization schemes.
result The solutions of gradient descent training are polyharmonic splines for certain initialization schemes.
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.
Theoretical analysis explains why models generalize after overfitting in modular addition.
problem Understanding why models generalize after overfitting in modular addition.
method Theoretical analysis and gradient descent behavior of two-layer quadratic networks and Transformers.
result Two-layer quadratic networks and simple Transformers generalize well after initially overfitting, indicating grokking.
Modern neural networks are typically trained in an over-parameterized regime where the parameters of the model far exceed the size of the training data. Such neural networks in principle have the capacity to (over)fit any set of labels including pure noise. Despite this, somewhat paradoxically, neural network models tr…
A new method uses extrapolation to train deep learning models with larger batch sizes.
problem Persistent degradation in performance when increasing batch size for deep learning.
method Proposes extrapolation to stabilize optimization trajectory and improve generalization.
result Demonstrates scaling to larger batch sizes while maintaining or surpassing state-of-the-art accuracy.
Recent work has demonstrated the effectiveness of gradient descent for directly recovering the factors of low-rank matrices from random linear measurements in a globally convergent manner when initialized properly. However, the performance of existing algorithms is highly sensitive in the presence of outliers that may …
New kernel improves MMDs with theoretical guarantees for gradient flows.
problem Non-smoothness of negative distance kernel in MMDs.
method Smoothed 1D absolute value function followed by fractional integral transform.
result Improved theoretical guarantees for Wasserstein gradient flows.
New perspective on SGD reveals short-range memory effects in deep learning.
problem Understanding the efficacy of stochastic gradient descent (SGD) in deep learning.
method Proposed that SGD is a discretization of an SDE driven by fractional Brownian motion (FBM).
result SGD stays longer in flat minima, favoring generalization.
FPG uses fractional calculus for efficient reinforcement learning with long-term memory.
problem High variance and inefficient sampling in standard policy gradient methods for long-term temporal modeling.
method Fractional Policy Gradients (FPG) incorporating Caputo fractional derivatives for power-law temporal correlations.
result Achieves asymptotic variance reduction of order O(t^(-alpha)) and sample efficiency gains.
New algorithms robustly estimate mean with near-optimal error rates.
problem Outlier robust mean estimation in high-dimensional data.
method Stability condition and iterative filtering algorithms.
result Optimal error rates with subgaussian rates for robust mean estimation.
The paper analyzes the maximum margin algorithm's performance on noisy data.
problem Analyzing the performance of maximum margin algorithm on noisy data.
method Finite-sample analysis of maximum margin algorithm applied to noisy data.
result The maximum margin algorithm can achieve nearly optimal population risk with sufficient over-parameterization.
Bayesian inference is a popular method to build learning algorithms but it is hampered by the fact that its key object, the posterior probability distribution, is often uncomputable. Expectation Propagation (EP) (Minka (2001)) is a popular algorithm that solves this issue by computing a parametric approximation (e.g: G…
We present a mathematical analysis of a non-convex energy landscape for robust subspace recovery. We prove that an underlying subspace is the only stationary point and local minimizer in a specified neighborhood under a deterministic condition on a dataset. If the deterministic condition is satisfied, we further show t…
Stochastic gradient descent algorithms for training linear and kernel predictors are gaining more and more importance, thanks to their scalability. While various methods have been proposed to speed up their convergence, the model selection phase is often ignored. In fact, in theoretical works most of the time assumptio…
We develop a variational framework for SDEs driven by fractional noise.
problem Capturing long-term dependencies in SDEs driven by fractional noise.
method Markov approximation of fractional Brownian motion, variational inference, neural networks.
result Efficient variational inference of posterior path measures for neural-SDEs.
Gradient Boosting Machine (GBM) introduced by Friedman is a powerful supervised learning algorithm that is very widely used in practice---it routinely features as a leading algorithm in machine learning competitions such as Kaggle and the KDDCup. In spite of the usefulness of GBM in practice, our current theoretical un…
In this paper we first identify a basic limitation in gradient descent-based optimization methods when used in conjunctions with smooth kernels. An analysis based on the spectral properties of the kernel demonstrates that only a vanishingly small portion of the function space is reachable after a polynomial number of g…
The question of how to parallelize the stochastic gradient descent (SGD) method has received much attention in the literature. In this paper, we focus instead on batch methods that use a sizeable fraction of the training set at each iteration to facilitate parallelism, and that employ second-order information. In order…
We consider the problem of sparsity-constrained M-estimation when both explanatory and response variables have heavy tails (bounded 4-th moments), or a fraction of arbitrary corruptions. We focus on the k-sparse, high-dimensional regime where the number of variables d and the sample size n are related through $…
Blind Descent avoids gradient issues, using a different learning approach.
problem Gradient issues like exploding and vanishing gradients.
method Does not use gradients to guide learning; instead, it is a more fundamental learning process.
result Gradient descent is a specific case of Blind Descent.
New method targets deep learning optima with heavy-tailed noise.
problem Heavy-tailed gradient noise biases SGDm optima.
method Fractional Underdamped Langevin Dynamics (FULD) for heavy-tailed noise.
result FULD targets Gibbs distribution matching optima of cost function.
Recent studies on diffusion-based sampling methods have shown that Langevin Monte Carlo (LMC) algorithms can be beneficial for non-convex optimization, and rigorous theoretical guarantees have been proven for both asymptotic and finite-time regimes. Algorithmically, LMC-based algorithms resemble the well-known gradient…
A new algorithm speeds up neural network training with less data.
problem Slow convergence of traditional neural network training methods.
method Decouples hidden layers and updates connections through bootstrapping, resampling, and linear regression.
result Empirically shows faster convergence with fewer data points.
In this note, we observe the behavior of gradient flow and discrete and noisy gradient descent in some simple settings. It is commonly noted that addition of noise to gradient descent can affect the trajectory of gradient descent. Here, we run some computer experiments for gradient descent on some simple functions, and…
New algorithm robustly optimizes data streams with heavy-tailed or infinite variance samples.
problem Optimizing data streams with heavy-tailed or infinite variance samples.
method Gradient quantile clipping for SGD, leveraging Markov chain connections.
result Algorithm converges to a concentrated distribution with high probability bounds.
Analyzes SGD dynamics on multi-class problems with exact expressions.
problem Analyzing SGD dynamics on multi-class problems.
method Developed a framework for analyzing training and learning rate dynamics using exact expressions.
result Exact expressions for risk and overlap with true signal in terms of ODEs.
Stochastic gradient descent on manifolds improves low-rank approximation.
problem Efficiently approximate large matrices with lower rank.
method Stochastic gradient descent on a manifold.
result Algorithm outperforms Euclidean space methods on Netflix Prize data.
A new method improves stochastic gradient descent for faster and more efficient estimation.
problem Efficient and fast parametric estimation methods.
method Projected stochastic gradient descent corrected by Fisher scoring.
result The method is faster and more efficient than traditional methods.
Gradient descent implicitly regularizes neural networks by penalizing large loss gradients.
problem How to optimize deep neural networks without explicit regularization.
method Backward error analysis to calculate implicit gradient regularization and demonstrate its effectiveness empirically.
result Implicit gradient regularization biases gradient descent toward flat minima, improving model robustness and test errors.
Derives Mirror Descent from gradient flow on a Riemannian manifold.
problem No specific problem stated; focuses on derivation.
method Derives Mirror Descent from gradient flow on a Riemannian manifold with a natural discretization.
result Generalizes Mirror Descent to non-Hessian metrics.
We consider the behavior of gradient flow and of discrete and noisy gradient descent. It is commonly noted that the addition of noise to the process of discrete gradient descent can affect the trajectory of gradient descent. In previous work, we observed such effects. There, we considered the case where the minima had …
Gradient descent variants improve phase retrieval accuracy.
problem Phase retrieval problem in high-dimensional spaces.
method Gradient descent, stochastic gradient descent, Langevin algorithm, dynamical mean-field theory.
result Stochastic variants of gradient descent achieve better generalization in phase retrieval.
Reparameterizes mirror descent as gradient descent for efficient sparse learning.
problem Efficiently training small sparse networks with mirror descent.
method Develops a framework to convert mirror descent updates into gradient descent updates on different parameters.
result Mirror descent can be reparameterized as gradient descent on modified parameters, facilitating standard backpropagation.
Gradient descent dynamics in nonconvex models explained with universality.
problem Understanding long-time behavior of nonconvex gradient descent.
method Developed a state evolution system for tracking gradient descent iterates.
result Gradient descent iterates are approximately independent of data and strongly incoherent with feature vectors.
Gravilon improves gradient descent for neural networks.
problem Improving efficiency and accuracy of gradient descent methods.
method Uses geometric modification of gradient step lengths.
result Promising experimental results on MNIST classification.
SGD reduces test error by decorrelating updates.
problem Improving generalization error in machine learning models.
method Derive a formula for generalization gap change due to SGD updates, compare to GD, and show decorrelation effect.
result SGD implicitly regularizes generalization error by decorrelating updates.
New adaptive step-size method for convex optimization without tuning.
problem Optimizing convex functions efficiently with stochastic gradients.
method Adapted Adaptive Gradient Descent Without Descent to stochastic setting.
result Stochastic gradient descent converges under various assumptions.
Natural gradient descent avoids the magic of model parametrization, leading to different optimization outcomes.
problem Understanding the impact of model parametrization on optimization and generalization in deep learning.
method Characterization of natural gradient flow in deep linear networks and nonlinear neural networks.
result Natural gradient descent fails to generalize in some cases, while gradient descent with the right architecture performs well.
Any gradient descent optimization requires to choose a learning rate. With deeper and deeper models, tuning that learning rate can easily become tedious and does not necessarily lead to an ideal convergence. We propose a variation of the gradient descent algorithm in the which the learning rate is not fixed. Instead, w…
Gradient descent biases towards stable rank networks for nearly-orthogonal data.
problem Understanding implicit bias in non-smooth neural networks trained by gradient descent.
method Analysis of two-layer ReLU and leaky ReLU networks trained by gradient descent on nearly-orthogonal data.
result Gradient descent biases towards networks with stable rank and uniform margin for nearly-orthogonal data.