AI Paper Atlas

Research map / Optimization / Optimizers

Stochastic Chebyshev Gradient Descent for Spectral Optimization

Insu Han, Haim Avron, Jinwoo Shin

NeurIPS 2018 · Optimization › Optimizers · approach: Randomized gradient estimation · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use unbiased randomized Chebyshev gradients with the optimal degree distributionSGD for matrix completion · MovieLens 1Mconvergence: about 2 times fasterImproved
use unbiased randomized Chebyshev gradients with the optimal degree distributionSVRG for matrix completion · MovieLens 10Mspeedup to RMSE 1.8: up to 6 times fasterImproved
use unbiased randomized Chebyshev gradientsSGD for Gaussian process regression · Natural sound datasetspeedup to RMSE 0.0375: 2 times fasterImproved
use unbiased randomized Chebyshev gradients instead of biased gradient estimatesSGD for Gaussian process regression · Szeged humid datasetconvergenceImproved
replace randomized degree sampling with a deterministic polynomial degree (ablation)SGD for matrix completion · MovieLens datasetsconvergenceHurt
replace the optimal degree distribution with a negative-binomial distributionSGD · MovieLens 1MconvergenceHurt
replace the optimal degree distribution with a Poisson distributionSGD · MovieLens 1MconvergenceHurt
increase the deterministic polynomial degree to N=30 (ablation)SGD-DET · MovieLens 1Mbias errorImproved
vary expected randomized polynomial degree between N=10 and N=30 (ablation)SGD · MovieLens 1MerrorNo clear effect
use SVRG instead of GDmatrix completion · MovieLens 1M and MovieLens 10Mconvergence speed: about 2 times faster; up to 6 times fasterImproved
increase the dataset dimension and parameter countSGD for matrix completion · MovieLens 10M versus MovieLens 1MconvergenceHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy