AI Paper Atlas

Research map / Optimization / Optimizers

Don't be so Monotone: Relaxing Stochastic Line Search in Over-Parameterized Models

Leonardo Galli, Holger Rauhut, Mark Schmidt

NeurIPS 2023 · Optimization › Optimizers · approach: Adam convergence and adaptive learning rates · 1 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 a nonmonotone rather than monotone stochastic line searchSGD/Adam with line search · Over-parameterized model experimentsconvergence speedImproved
Use PoNoS instead of SLSSGD with line search · Image classification experimentstest accuracyImproved
Use PoNoS instead of SGD or AdamImage classification experiments · Image classification experimentstrain lossImproved
Use PoNoS instead of SGDImage classification training · Image classification experimentstotal computational timeImproved
Combine the nonmonotone line search with the Polyak initial step sizePoNoS · Mushroomsiterations to reach train loss 10^{-4}: 6 times the iterationsImproved
Use PoNoS_prec instead of other tested methodsPreconditioned transformer training · PTB and Wikitext2train loss convergence speedMixed
Use the new resetting technique (ablation)PoNoS · Image classification experimentsnumber of backtracks: 0Improved
Remove the new resetting technique (ablation)PoNoS_reset0 · Image classification experimentstrain lossNo clear effect
Compute a new step only every 2 iterations (ablation)PoNoS_reset0 · Image classification experimentstrain lossHurt
Use the alternative Zhang resetting initial step sizes (ablation)Nonmonotone line search · Image classification experimentstrain lossHurt
Use the smoothed Polyak initial step size instead of the non-smoothed version (ablation)SPS · MLP and FashionMNISTconvergence speedHurt
Set the line-search parameter c to 0.1 rather than a larger value (ablation)Stochastic line search · Experimentsstability: 0.1Hurt

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