We shed new insights on the two commonly used updates for the online -PCA problem, namely, Krasulina's and Oja's updates. We show that Krasulina's update corresponds to a projected gradient descent step on the Stiefel manifold of the orthonormal -frames, while Oja's update amounts to a gradient descent step using…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,742 papers · 148 categories
Trend · papers per month
48 results for “gradient step update”
Proposes a method to balance tasks in multitask learning with a single gradient step update.
problem Balancing tasks in multitask learning to avoid imbalance.
method Gradient-based meta-learning to balance tasks at the gradient level, training shared and task-specific layers separately.
result Achieves state-of-the-art performance on various multitask computer vision problems.
New insights into neural network feature learning through multi-step gradient descent.
problem Understanding feature learning in two-layer neural networks with limited width.
method Characterization of feature learning through two steps of gradient descent with specific step sizes.
result The second step of gradient descent reveals multiple learned directions, not limited to a single direction as in the first step.
GD converges in unstable regimes, even with oscillatory behavior.
problem Understanding convergence of GD in unstable regimes.
method Analysis of two-step gradient updates.
result Characterization of local conditions for convergence.
Automatic, Dynamic, and Nearly Optimal Learning Rate Specification by Local Quadratic Approximationstat.ML
A new method automatically and dynamically sets learning rates in deep learning.
problem Determining the appropriate learning rate in deep learning tasks is challenging and often subjective.
method Local Quadratic Approximation (LQA) to automatically and dynamically set learning rates.
result The proposed method leads to nearly optimal learning rates in a computationally efficient way.
A major challenge in current optimization research for deep learning is to automatically find optimal step sizes for each update step. The optimal step size is closely related to the shape of the loss in the update step direction. However, this shape has not yet been examined in detail. This work shows empirically that…
Convergence of the Stochastic Heavy Ball Method With Approximate Gradients and/or Block Updatingmath.OC
Paper proves SHB convergence with biased gradients and approximate step sizes.
problem Establishing convergence of SHB with biased gradients and approximate step sizes.
method Generalizes SHB convergence conditions for biased gradients, approximate step sizes, and block updating.
result Proves convergence of SHB with new conditions for biased gradients and approximate step sizes.
RSO uses random weight perturbations to train deep networks without gradients.
problem Training deep neural networks efficiently and without gradient information.
method RSO is a gradient-free Markov Chain Monte Carlo approach that updates weights based on mini-batch loss reduction.
result RSO achieves high accuracy (99.1% on MNIST) with significantly fewer updates than traditional methods.
A new method for machine learning updates reduces complexity and improves robustness.
problem Stochastic gradient updates are inefficient and sensitive to feature scaling.
method Incremental Gauss-Newton Descent (IGND) reduces the need for matrix operations and improves robustness.
result IGND improves robustness to sensitivity scaling and can be competitive with common stochastic optimizers.
W-Flow generates images in one step, faster and better than multi-step methods.
problem Efficiently generating images from a simple reference distribution to a target data distribution.
method W-Flow uses Wasserstein gradient flows to transform the reference distribution to the target distribution in a single step, trained with Sinkhorn divergence.
result W-Flow achieves state-of-the-art results in ImageNet 256256 generation with improved mode coverage and faster sampling.
The rise of deep learning in recent years has brought with it increasingly clever optimization methods to deal with complex, non-linear loss functions. These methods are often designed with convex optimization in mind, but have been shown to work well in practice even for the highly non-convex optimization associated w…
New method improves convergence of RL meta-learning.
problem Improving convergence in model-agnostic meta-reinforcement learning.
method Proposes Stochastic Gradient Meta-Reinforcement Learning (SG-MRL) to find -first-order stationary points.
result Derives iteration and sample complexity for SG-MRL.
We present a predictor-corrector framework, called PicCoLO, that can transform a first-order model-free reinforcement or imitation learning algorithm into a new hybrid method that leverages predictive models to accelerate policy learning. The new "PicCoLOed" algorithm optimizes a policy by recursively repeating two ste…
This paper investigates different vector step-size adaptation approaches for non-stationary online, continual prediction problems. Vanilla stochastic gradient descent can be considerably improved by scaling the update with a vector of appropriately chosen step-sizes. Many methods, including AdaGrad, RMSProp, and AMSGra…
Stagewise boosting improves gradient boosting for distributional regression.
problem Vanishing gradient in gradient boosting for distributional regression leads to suboptimal models.
method Proposes a stagewise boosting-type algorithm for distributional regression, combining stagewise regression ideas with gradient boosting and incorporating a novel regularization method, correlation filtering.
result The proposed algorithm provides better results, especially for complex distributions, by reducing the risk of being trapped in a local optimum.
GENIE balances domain-invariant feature learning and gradient alignment for improved DG performance.
problem Domain Generalization (DG) overfitting to domain-specific features
method GENIE (Generalization-ENhancing Iterative Equalizer) optimizer
result Prevents a small subset of parameters from dominating optimization, promoting domain-invariant feature learning
The paper analyzes RLVR's training dynamics, proving convergence depends on aligning update direction with Gradient Gap.
problem Understanding why RLVR works and its limitations.
method Analysis of RLVR's training process at trajectory and token levels, introducing Gradient Gap.
result Convergence depends on aligning update direction with Gradient Gap, with a sharp step-size threshold.
Gradient equilibrium improves online learning performance without requiring sublinear regret.
problem Achieving sublinear regret in online learning.
method Gradient equilibrium: average of gradients converges to zero.
result Gradient equilibrium can be achieved by standard online learning methods.
Federated learning improves by unbiased gradient aggregation and controllable meta updating.
problem Gradient biases and inconsistency between target and optimization objectives in federated averaging.
method Unbiased gradient aggregation with keep-trace gradient descent and gradient evaluation strategy, controllable meta updating with small data samples.
result Faster convergence and higher accuracy with different network architectures in various FL settings.
Min-max formulations have attracted great attention in the ML community due to the rise of deep generative models and adversarial methods, while understanding the dynamics of gradient algorithms for solving such formulations has remained a grand challenge. As a first step, we restrict to bilinear zero-sum games and giv…
Designing deterministic denominators for SGLD stabilizes large drifts.
problem Stabilizing large drifts in SGLD
method Using state-dependent envelopes and empirical quantiles for activation thresholds
result Proxy-quantile denominators are close to oracle-score behavior and improve deterministic taming choices
An EM Gradient Algorithm for Mixture Models with Components Derived from the Manly Transformationstat.ML
A new EM gradient algorithm for mixture models with skewed components.
problem Fitting mixture models with skewed components derived from the Manly transformation.
method Proposes an alternative EM gradient algorithm using Newton's method for better parameter updates.
result Shows improved convergence and parameter estimation compared to the Nelder-Mead optimization.
Proposes a semi-implicit back propagation method for neural networks.
problem Challenges in training neural networks, especially gradient vanishing and small step sizes.
method Proposes a semi-implicit back propagation method using error back propagation and proximal methods.
result The proposed method leads to better performance in terms of loss decreasing and training/validation accuracy compared to SGD and ProxBP.
Gradient-based methods for games suffer from discrete update steps that cause drift, affecting performance.
problem Gradient-based methods for two-player games suffer from drift due to discrete update steps.
method Derived modified continuous dynamical systems to closely follow the discrete dynamics of games.
result Identified distinct components of discretization drift that can alter or destabilize game performance.
SFPO optimizes LLM reasoning by repositioning before updating, improving stability and efficiency.
problem Noisy gradients from low-quality rollouts cause instability and inefficient exploration in on-policy RL algorithms.
method Decomposes each step into three stages: a short fast trajectory, repositioning, and slow correction, preserving the objective and rollout process unchanged.
result SFPO consistently improves stability, reduces rollouts, and accelerates convergence, outperforming GRPO on math reasoning benchmarks.
Many applications in signal processing benefit from the sparsity of signals in a certain transform domain or dictionary. Synthesis sparsifying dictionaries that are directly adapted to data have been popular in applications such as image denoising, inpainting, and medical image reconstruction. In this work, we focus in…
Gradient-EM Bayesian meta-learning accelerates adaptation with reduced computation and improved robustness.
problem Efficient and robust adaptation to new tasks with uncertainty assessment.
method Extends Bayesian meta-learning with gradient-EM algorithm, decoupling inner-update from meta-update.
result Improves accuracy with less computation cost and enhanced robustness to uncertainty.
AEGD optimizes non-convex functions with dynamic energy updates.
problem Optimizing non-convex functions efficiently and robustly.
method Adaptive Gradient Descent (AEGD) with a dynamically updated energy variable.
result AEGD achieves energy-dependent convergence rates for both convex and non-convex objectives.
Unbiased Gradient Estimation in Unrolled Computation Graphs with Persistent Evolution Strategiescs.LG
PES method reduces bias in gradient estimation for unrolled graphs.
problem High variance and bias in gradient estimation for unrolled computation graphs.
method Divide graph into unrolls, apply ES update, accumulate correction terms.
result PES provides unbiased, low-variance gradient estimates.
New actor-critic method reduces sample complexity for reinforcement learning.
problem Improving sample complexity for actor-critic algorithms in reinforcement learning.
method Integrates Monte Carlo rollouts into policy search steps for better control over bias.
result Established sample complexity for actor-critic algorithms with policy gradient.
The paper deals with learning probability distributions of observed data by artificial neural networks. We suggest a so-called gradient conjugate prior (GCP) update appropriate for neural networks, which is a modification of the classical Bayesian update for conjugate priors. We establish a connection between the gradi…
AB dynamically scales gradients to mitigate asynchronous training delays.
problem Gradient delay in asynchronous training reduces model performance.
method Adaptive Braking (AB) dynamically scales gradients based on alignment.
result AB enables training with up to 32 update steps of delay without accuracy loss.
A new approach RA improves stochastic optimization by executing multiple steps between subsample updates.
problem Improving the efficiency and effectiveness of stochastic optimization methods.
method Developed Retrospective Approximation (RA) which executes multiple steps between subsample updates using a deterministic solver.
result RA achieves almost sure and consistency under weak conditions and optimizes iteration and oracle complexity.
A new method for robust training under label noise using weighted gradient descent.
problem Overfitting to noisy examples in machine learning.
method Exponentiated gradient reweighting for flexible handling of noisy data.
result Improved generalization in noisy classification and PCA problems.
Optimistic method adapted for faster convex-concave min-max problems.
problem Solving convex-concave min-max optimization problems efficiently.
method Adaptive, line search-free second-order methods combining optimistic updates and second-order information.
result Achieves optimal convergence rate without line search or backtracking.
LatentTrack generates model parameters online for nonstationary data.
problem Online probabilistic prediction under nonstationary dynamics.
method Sequential neural architecture with latent filtering and amortized inference.
result Consistently lower negative log-likelihood and mean squared error than baselines.
New SPS variant improves non-smooth optimization without small gradients.
problem Improving non-smooth optimization without small gradients.
method Safeguarded Stochastic Polyak Step Size (SPS) for non-smooth optimization.
result Rigorous convergence guarantees for non-smooth convex optimization without strong assumptions.
RELTA-SGLD: Relative-Growth Localized Taming for Nonconvex Stochastic-Gradient Langevin Learningstat.ML
RELTA-SGLD stabilizes nonconvex SGLD updates with a lighter taming scheme.
problem Stabilizing superlinear stochastic-gradient updates in nonconvex optimization.
method Threshold-based taming with relative-growth principle for stability.
result Polynomial moment stability and first-order stationary accuracy in nonconvex SGLD.
Muon replaces matrix gradient with polar factor, optimizing flat spectrum updates
problem Optimization bias in matrix updates
method Using polar factor of gradient
result Muon update maximizes entropy among bounded updates
Accelerates policy optimization in RL with optimistic and adaptive updates.
problem Improving policy optimization methods in reinforcement learning.
method Integrates foresight into policy improvement step via optimistic and adaptive updates.
result Designs an optimistic policy gradient algorithm, adaptive via meta-gradient learning.
This paper proposes a novel proximal-gradient algorithm for a decentralized optimization problem with a composite objective containing smooth and non-smooth terms. Specifically, the smooth and nonsmooth terms are dealt with by gradient and proximal updates, respectively. The proposed algorithm is closely related to a p…
Fewer data weight updates lead to faster convergence in machine learning models.
problem Improving robustness of machine learning models through data mixing.
method Analyzing convergence behavior of data mixing with a finite number of inner steps.
result The optimal number of inner steps scales with the budget and type of gradients used.
A new PGA algorithm ensures stable, robust, and noise-immune solutions for non-negative inverse problems.
problem Stable convergence and suboptimal solutions in inverse problems due to negative values and high sensitivity to hyperparameters.
method A novel multiplicative update proximal gradient algorithm (SSO-PGA) that enforces non-negativity and boundedness through a learnable sigmoid-based operator.
result Significantly surpasses traditional PGA and other state-of-the-art algorithms in performance and stability.
Drop-Muon updates only some layers, speeding up training.
problem Conventional deep learning optimizers update all layers at once, which can be inefficient.
method Drop-Muon updates only a subset of layers per step, with randomized schedules.
result Drop-Muon achieves up to 1.4x faster training time with similar accuracy.
Paper tackles efficient policy gradient estimation from off-policy data.
problem Estimating policy gradients from off-policy data is challenging and inefficient.
method Derives asymptotic lower bounds, proposes a meta-algorithm with 3-way robustness, and establishes convergence guarantees.
result Meta-algorithm achieves the lower bound on mean-squared error without parametric assumptions.
An online decision-making algorithm using stochastic gradient descent for big data.
problem Efficiently updating decision rules in online decision making with big data.
method Stochastic gradient descent for online updates, asymptotic normality of estimators.
result Asymptotic normality of parameter and value estimators, enabling statistical inference.
EP learns like BPTT but with local weight updates.
problem Existing EP lacks a local time learning rule.
method C-EP updates weights simultaneously with neuron dynamics.
result C-EP follows BPTT gradients and performs well.
New TD algorithms stabilize RL tasks by reformulating updates into fixed point equations.
problem TD learning's sensitivity to step size specification.
method Implicit TD algorithms reformulate TD updates into fixed point equations.
result Implicit TD algorithms are more stable and less sensitive to step size.