AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Deep Bilevel Learning

Simon Jenni, Paolo Favaro

ECCV 2018 · Optimization › Implicit bias and generalization · approach: Stability-based generalization analysis · 113 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
add L1 normalization of the mini-batch weights (ablation)Bilevel (CifarNet) · CIFAR-10 clean labelsclean test accuracy: 74.32 -> 75.52Improved
add L1 normalization of the mini-batch weights (ablation)Bilevel (Inception) · CIFAR-10 clean labelsclean test accuracy: 77.07 -> 87.78Improved
force identical label distribution across compared mini-batches (ablation)Bilevel (Inception) · CIFAR-10 50% random label noiseclean test accuracy: 27.20 -> 73.08Improved
force identical label distribution across compared mini-batches (ablation)Bilevel (CifarNet) · CIFAR-10 clean labelsclean test accuracy: 68.19 -> 75.52Improved
compute separate weights per layer instead of one per mini-batch (ablation)Bilevel (CifarNet) · CIFAR-10 50% random label noiseclean test accuracy: 47.62 -> 49.62Improved
compute separate weights per layer instead of one per mini-batch (ablation)Bilevel (Inception) · CIFAR-10 clean labelsclean test accuracy: 87.78 -> 85.25Hurt
allow unequal dropout across compared mini-batches (ablation)Bilevel (Inception) · CIFAR-10 50% random label noiseclean test accuracy: 73.08 -> 75.71Improved
increase number of compared mini-batches (ablation)Bilevel · CIFAR-10 (CifarNet and Inception)clean test accuracyMixed
increase validation ratio (ablation)Bilevel · CIFAR-10test accuracyNo clear effect
vary the hyper-parameter mu (ablation)Bilevel · CIFAR-10test accuracyNo clear effect
use bilevel optimization instead of standard SGDPreAct ResNet-18 · CIFAR-10 (40% label noise)test accuracy: 69.6% -> 87.0%Improved
use bilevel optimization instead of standard SGDPreAct ResNet-18 · CIFAR-100 (40% label noise)test accuracy: 44.9% -> 59.8%Improved

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