AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Deep Metric Learning with BIER: Boosting Independent Embeddings Robustly

Michael Opitz, Georg Waltner, Horst Possegger, Horst Bischof

TPAMI 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 165 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 boosting-based reweighting on top of Activation Loss initialization (ablation)BIER 96-160-256 · CUB-200-2011Recall@1: 53.93 -> 55.33Improved
add Activation Loss initialization with split groups and no boosting (ablation)Baseline-512 · CUB-200-2011Recall@1: 51.76 -> 53.93Improved
use groups of unequal size 96-160-256 instead of 170-171-171 (ablation)BIER · CUB-200-2011Recall@1: 54.76 -> 55.33Improved
apply BIER to triplet loss (ablation)Triplet-512 · CUB-200-2011Recall@1: 50.12 -> 53.31Improved
apply BIER to contrastive loss (ablation)Contrastive-512 · CUB-200-2011Recall@1: 50.62 -> 53.8Improved
apply BIER to binomial deviance loss (ablation)Binomial-Deviance-512 · CUB-200-2011Recall@1: 51.76 -> 55.33Improved
increase groups from 4 to 5 at 512-d embedding (ablation)BIER · CUB-200-2011Recall@1: 55.62 -> 54.9Hurt
apply BIER to 1024-d embedding (ablation)Baseline-1024 · CUB-200-2011Recall@1: 52.89 -> 55.99Improved
apply BIER to 384-d embedding (ablation)Baseline-384 · CUB-200-2011Recall@1: 51.57 -> 54.66Improved
use Activation Loss initialization instead of Glorot random initialization (ablation)BIER · CUB-200-2011Recall@1: 54.41 -> 55.33Improved
use Activation Loss initialization instead of orthogonal initialization (ablation)BIER · CUB-200-2011Recall@1: 54.58 -> 55.33Improved
raise learning rate to 1e-5 without auxiliary loss (ablation)BIER · CUB-200-2011Recall@1: 55.3 -> 52.3Hurt

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