AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Diversity With Cooperation: Ensemble Methods for Few-Shot Classification

Nikita Dvornik, Cordelia Schmid, Julien Mairal

ICCV 2019 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Meta-learning analysis and benchmarks · 214 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 cooperation penalty (symmetrized KL on non-ground-truth probabilities) to a small ensemble of 5 networksindependent ensemble · miniImageNet 5-shotaccuracy (%): 80.02 -> 80.60Improved
add diversity penalty to a large ensemble of 20 networksindependent ensemble · miniImageNet 5-shotaccuracy (%): 80.57 -> 81.18Improved
add robust strategy (cooperation plus network dropping, dropout, different augmentations) to 20 networksindependent ensemble · miniImageNet 1-shotaccuracy (%): 62.06 -> 63.95Improved
add cooperation penalty to a 20-network ensembleindependent ensemble · miniImageNet 5-shotaccuracy (%): 80.57 -> 80.80No clear effect
distill 20-network robust ensemble into one network using extra unlabeled COCO imagesrobust ensemble · miniImageNet 5-shotaccuracy (%): 81.59 -> 81.19Hurt
add extra unannotated images to distillation (ablation)robust-dist, 20 networks · miniImageNet 5-shotaccuracy (%): 80.63 -> 81.19Improved
add extra unannotated COCO images to distillation on CUB (ablation)robust-dist, 20 networks · CUB 5-shotaccuracy (%): 83.57 -> 83.21Hurt
use conditional non-ground-truth probabilities instead of full softmax probabilities in cooperation penalty (ablation)cooperation ensemble, 5 networks · miniImageNet 5-shotaccuracy (%): 79.79 -> 80.60Improved
use symmetrized KL instead of L2 for cooperation (ablation)cooperation ensemble, 5 networks · miniImageNet 5-shotaccuracy (%): 80.14 -> 80.72Improved
use negative KL instead of cosine for diversity (ablation)diversity ensemble, 5 networks · miniImageNet 5-shotaccuracy (%): 80.82 -> 79.81Hurt
use cosine instead of L2 for diversity (ablation)diversity ensemble, 5 networks · miniImageNet 5-shotaccuracy (%): 80.54 -> 80.82No clear effect
use learned weights for class prototypes instead of mean centroid (ablation)mean centroid classifier · miniImageNet 5-shotaccuracy (%): 77.28 -> 77.01No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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