AI Paper Atlas

Research map / Continual learning / Class-incremental

FeCAM: Exploiting the Heterogeneity of Class Distributions in Exemplar-Free Continual Learning

Dipam Goswami, Yuyang Liu, Bartłomiej Twardowski, Joost van de Weijer

NeurIPS 2023 · Continual learning › Class-incremental · approach: Exemplar-efficient class-incremental learning · 23 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
replace Euclidean NCM with per-class Mahalanobis (FeCAM)Eucl-NCM · CIFAR-100 T=5avg accuracy (%): 64.8 -> 70.9Improved
use per-class covariance instead of a common covariance (ablation)FeCAM · TinyImageNet T=5avg accuracy (%): 56.0 -> 59.6Improved
use FeCAM instead of FeTrILFeTrIL · ImageNet-Subset T=5avg accuracy (%): 73.1 -> 78.3Improved
use FeCAM instead of L2P and NCM with pretrained ViTL2P / NCM · CoRe50accuracy (%): 78.3 -> 89.9Improved
use FeCAM without exemplars vs exemplar-based DERDER · CIFAR-100 T=5last accuracy (%): 66.2 -> 62.1Hurt
add Tukey transformation to Euclidean NCM (ablation)Euclidean NCM · CIFAR-100 T=5avg accuracy (%): 64.8 -> 66.6Improved
add covariance shrinkage to full Mahalanobis (ablation)Mahalanobis · CIFAR-100 T=5avg accuracy (%): 29.7 -> 59.3Improved
add Tukey transformation to full Mahalanobis with shrinkage (ablation)Mahalanobis · CIFAR-100 T=5avg accuracy (%): 59.3 -> 62.8Improved
add correlation normalization (ablation)Mahalanobis with Tukey and shrinkage · CIFAR-100 T=5avg accuracy (%): 62.8 -> 70.9Improved
use full covariance instead of diagonal (ablation)FeCAM · CIFAR-100 T=5avg accuracy (%): 66.9 -> 70.9Improved
use Bayes Mahalanobis classifier instead of sampling features and training a linear classifier (ablation)FeCAM · CIFAR-100 T=5avg accuracyImproved
use FeCAM instead of FeTrIL with small first task (20 classes)FeTrIL · ImageNet-Subsetavg accuracy (%): 63.1 -> 66.4Improved

Later papers that compared with this method

7 later accepted papers report FeCAM as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.

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