AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Single Loop Gaussian Homotopy Method for Non-convex Optimization

Hidenori Iwakiri, Yuhang Wang, Shinji Ito, Akiko Takeda

NeurIPS 2022 · Optimization › Nonconvex and smooth optimization · approach: Complexity of nonconvex first-order methods

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 single-loop homotopy ZOSLGHr instead of double-loop ZOGradOptZOSLGH · CIFAR-10 per-image attacksuccess rate: 65% -> 93%Improved
Use ZOSLGHr instead of ZOGradOptZOSLGH · CIFAR-10 per-image attackavg iterations to first success: 6789 ± 1901 -> 4979 ± 756Improved
Use ZOSLGHr instead of ZOSGDZOSLGH · CIFAR-10 per-image attacksuccess rate: 88% -> 93%Improved
Use ZOSLGHr instead of ZOSGDZOSLGH · CIFAR-10 per-image attackavg iterations to first success: 835 ± 1238 -> 4979 ± 756Hurt
Use ZOSLGHr instead of ZOSGDZOSLGH · CIFAR-10 per-image attackavg total loss: 27.70 ± 74.80 -> 14.26 ± 54.61Improved
Use ZOSLGHr instead of ZOAdaMMZOSLGH · CIFAR-10 per-image attackavg L2 distortion (successful): 0.050 ± 0.055 -> 0.246 ± 0.178Hurt
Use ZOSLGHd (derivative-based t update) instead of ZOSLGHrZOSLGH · CIFAR-10 per-image attackavg L2 distortion (successful): 0.246 ± 0.178 -> 0.150 ± 0.084Improved
Use ZOSLGHd instead of ZOSLGHrZOSLGH · CIFAR-10 per-image attacksuccess rate: 93% -> 92%No clear effect
Use ZOSLGHd instead of ZOSLGHrZOSLGH · CIFAR-10 per-image attackavg total loss: 14.26 ± 54.61 -> 16.49 ± 58.69No clear effect
Use ZOSLGHd instead of ZOSLGHrZOSLGH · CIFAR-10 per-image attackcomputational time per iterationHurt
Use ZOAdaMM instead of ZOSGDZO baselines · CIFAR-10 per-image attacksuccess rate: 88% -> 85%Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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