AI Paper Atlas

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

Reconstruct & Crush Network

Erinc Merdivan, Mohammad Reza Loghmani, Matthieu Geist

NeurIPS 2017 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 2 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
use RCN instead of a standard CNNautoencoder classifier · CIFAR-10 unseen shiptrue negative rate: 0.755 -> 0.793Improved
use RCN instead of a standard CNNautoencoder classifier · CIFAR-10 automobiletrue positive rate: 0.74 -> 0.81Improved
use RCN instead of class-balanced CNN-reducedautoencoder classifier · CIFAR-10 automobiletrue positive rate: 0.82 -> 0.81No clear effect
use RCN instead of CNN-reducedautoencoder classifier · CIFAR-10 unseen shiptrue negative rate: 0.638 -> 0.793Improved
balance positive and negative training sets (CNN-reduced vs CNN)CNN · CIFAR-10 unseen shiptrue negative rate: 0.755 -> 0.638Hurt
use RCN instead of CNNautoencoder classifier · CIFAR-100 platestrue positive rate: 0.718 -> 0.861Improved
use RCN instead of CNNautoencoder classifier · CIFAR-100 tulipstrue negative rate: 0.81 -> 0.853Improved
use RCN instead of PU-specific methods with 5% positives in unlabeled setRCN · Amazon reviews PUF-measure: 0.92 -> 0.90Hurt
raise positive share of unlabeled set from 5% to 50%RCN · Amazon reviews PUF-measure: 0.90 -> 0.83Hurt
raise positive share of unlabeled set from 5% to 50%NB-SVM · Amazon reviews PUF-measure: 0.92 -> 0.86Hurt
raise positive share of unlabeled set from 5% to 50%Roc-EM · Amazon reviews PUF-measure: 0.91 -> 0.90No clear effect
use RCN energy as response scorerGRU autoencoder · Facebook bAbI tasks 1 and 2accuracy: 100% accuracyNo 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