AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Instance-Dependent Generalization Bounds via Optimal Transport

Songyan Hou, Parnian Kassraie, Anastasis Kratsios, Andreas Krause, Jonas Rothfuss

JMLR 2023 · Optimization › Implicit bias and generalization · approach: Stability-based generalization analysis

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 partitioned local-Lipschitz bound instead of the global Lipschitz boundpartition bound · classification error bound, N=2560generalization bound: 18.407 -> 1.447Improved
use a partitioned local-Lipschitz bound instead of the global Lipschitz boundpartition bound · classification error bound, N=81920generalization bound: 2.856 -> 0.478Improved
use partition bound instead of the uniform Rademacher boundpartition bound · classification error bound, N=2560generalization bound: 39.545 -> 1.447Improved
use partition bound instead of the uniform Rademacher boundpartition bound · classification error bound, N=81920generalization bound: 16.738 -> 0.478Improved
increase the number of training points from 2560 to 81920partition bound · classification error boundgeneralization bound: 1.447 -> 0.478Improved
partition the input domain into parts finer and finer (ablation)partition bound · regression and classificationgeneralization bound: bound values initially decrease and, as the partitioning becomes much finer, increase againMixed
increase network size by a factor 1000 in parameters (ablation)partition bound · regression and classificationgeneralization bound: only a minor effect on the value of the boundNo clear effect

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