arXiv research

A method to train deep neural networks on noisy labeled data.

Unified framework for analyzing batch updating methods with noisy gradients.

The study examines backward compatibility issues in ML systems, especially with noisy data.

New bounds for noisy iterative algorithms reduce overfitting risk.

A new method for robust training under label noise using weighted gradient descent.

Framework corrects noisy labels to improve DNN performance.

The paper analyzes generalization of noisy, iterative algorithms using maximal leakage.

A novel framework optimizes experience replay for reinforcement learning.

Improved neural network inference with eigenvalue correction.

SFPO optimizes LLM reasoning by repositioning before updating, improving stability and efficiency.

Develops an off-policy method for improving chatbots with noisy rewards.

Co-teaching improves deep neural networks' robustness against noisy labels.

PIE-PINN estimates elastic properties from noisy, low-res displacement data.

New method tackles noisy and incomplete observations in reinforcement learning.

JoCoR improves deep learning with noisy labels by reducing network diversity.

MSLG generates soft labels to improve DNN performance on noisy datasets.

A new numerical scheme approximates nonlinear filtering densities for noisy and partial measurements.

Learning with noisy labels is one of the hottest problems in weakly-supervised learning. Based on memorization effects of deep neural networks, training on small-loss instances becomes very promising for handling noisy labels. This fosters the state-of-the-art approach "Co-teaching" that cross-trains two deep neural ne…

Gradient filters track moving parameters under noisy data and misspecification.

PBA solves root-finding problems with noisy responses, converging slower than stochastic approximation.

Locally private reinforcement learning protects individual environments from reverse engineering.

New method for asynchronous stochastic approximation converges in reinforcement learning.

DONE algorithm optimizes unknown functions with noisy measurements.

New algorithms minimize noisy, irregular functions without gradients.

A new method for efficient neural network fine-tuning using queryable low-rank update atoms.

Study evaluates GP metamodels and sequential designs for noisy level set estimation.

SAUNA filters out noisy samples to boost RL performance.

Study tackles nonlinear factor models with unknown monotone links from incomplete and noisy data.

This paper proposes a new estimation algorithm for the parameters of an HMM as to best account for the observed data. In this model, in addition to the observation sequence, we have \emph{partial} and \emph{noisy} access to the hidden state sequence as side information. This access can be seen as "partial labeling" of …

We develop methods for parameter estimation in settings with large-scale data sets, where traditional methods are no longer tenable. Our methods rely on stochastic approximations, which are computationally efficient as they maintain one iterate as a parameter estimate, and successively update that iterate based on a si…

Improves scalability and robustness of dynamic graph clustering.

New algorithm detects and discards faulty updates in federated learning.

A new method approximates expected empirical loss for stochastic deep learning tasks.

New algorithm RBO improves RL in noisy environments.

Neural networks learn clean data patterns first, then noisy data, leading to improved performance initially but deteriorating later.

A framework for navigating environments with spatially correlated obstacles and uncertain blockage status.

Approximate Message Passing (AMP) has been shown to be a superior method for inference problems, such as the recovery of signals from sets of noisy, lower-dimensionality measurements, both in terms of reconstruction accuracy and in computational efficiency. However, AMP suffers from serious convergence issues in contex…

Paper tackles federated linear bandit learning with AirComp for noisy channels.

Generalizes root-finding algorithm for noisy responses.

Efficient learning with robust gradient descent reduces resource usage.

Aims to improve neural network training with weak labels by weighting updates based on a confidence network.

The question of how to incorporate curvature information in stochastic approximation methods is challenging. The direct application of classical quasi- Newton updating techniques for deterministic optimization leads to noisy curvature estimates that have harmful effects on the robustness of the iteration. In this paper…

Method learns dynamics from noisy partial observations.

Online learning framework for inverse optimization improves decision-making in noisy data.

Tensor-EM method learns MoLDS from complex, noisy data.

DaRE forests enable efficient data deletion from random forests.

Stochastic Gradient Descent (SGD) has become one of the most popular optimization methods for training machine learning models on massive datasets. However, SGD suffers from two main drawbacks: (i) The noisy gradient updates have high variance, which slows down convergence as the iterates approach the optimum, and (ii)…

A-FADMM improves FL scalability and privacy via wireless channel perturbations and interference.