AI Paper Atlas

Research map / Optimization / Optimizers

Better Parameter-Free Stochastic Optimization with ODE Updates for Coin-Betting

Keyi Chen, John Langford, Francesco Orabona

AAAI 2022 · Optimization › Optimizers · approach: SGD momentum and step-size schedules · 3 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 CODE instead of CoinParameter-free coin-betting optimization · 17 regression datasets, best fixed learning ratemean normalized absolute loss: CODE wins Coin by ~0.1984Improved
Use CODE instead of AdaGradOptimization on 17 regression datasets · 17 regression datasets, best fixed learning ratemean normalized absolute loss: 1.1227 -> 0.7825Improved
Use CODE instead of AdaGradOptimization on 21 classification datasets · 21 classification datasets, best fixed learning ratemean normalized 0-1 loss: 0.1778 -> 0.1872Hurt
Use CODE instead of AdaGradOptimization on 17 regression datasets · 17 regression datasets, tuned learning ratesmean normalized absolute loss: 0.7085 -> 0.7825Hurt
Use CODE instead of AdaGradOptimization on 21 classification datasets · 21 classification datasets, tuned learning ratesmean normalized 0-1 loss: 0.1694 -> 0.1872Hurt
Replace tuned learning-rate baselines with parameter-free CODEOptimization on regression and classification datasets · 17 regression and 21 classification datasetsnormalized test loss: ~0.0740 on regression problems; ~0.0178 on classification problemsNo clear effect
Use CODE rather than step-size-sensitive SGD, Adam, or AdaGradSynthetic absolute-loss regression · Synthetic dataset, noiseless and noisy regimesiterations to reach 0.05 suboptimality gapImproved
Use CODE rather than CoinParameter-free optimization on synthetic regression · Synthetic dataset, noiseless regimeiterations to reach 0.05 suboptimality gapImproved

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