Stochastique Gradient Descent

Ou descente de gradient stochastique en français.

(à venir)

Lectures

• Adam: A Method for Stochastic Optimization

• HOGWILD!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent

• Sparse Online Learning via Truncated Gradient

• Stabilized Sparse Online Learning for Sparse Data

• On Perturbed Proximal Gradient Algorithms

• Large Margin Classification Using the Perceptron Algorithm

• Scaling Up Stochastic Dual Coordinate Ascent

• Stochastic Majorization-Minimization Algorithms for Large-Scale Optimization

• Dual Principal Component Pursuit

• Parallelizing Stochastic Gradient Descent for Least Squares Regression: Mini-batching, Averaging, and Model Misspecification

• Compact Convex Projections

• Optimization Methods for Large-Scale Machine Learning

• k-SVRG: Variance Reduction for Large Scale Optimization

• Accelerating Stochastic Gradient Descent using Predictive Variance Reduction

• Gradients without Backpropagation

• Large Batch Optimization for Deep Learning: Training BERT in 76 minutes