AI Paper Atlas

Research map / Continual learning / Online and replay-based

GDumb: A Simple Approach that Questions Our Progress in Continual Learning

Ameya Prabhu, Philip H. S. Torr, Puneet K. Dokania

ECCV 2020 · Continual learning › Online and replay-based · approach: Plug-and-play online training enhancements

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 GDumb instead of ER-MIRcontinual learner · Split-CIFAR10 A1 (k=1000)accuracy: 47.6 -> 61.3Improved
use GDumb instead of ER-MIRcontinual learner · Split-MNIST A1 (k=500)accuracy: 87.6 -> 91.9Improved
use GDumb instead of GMED (MIR)continual learner · Split-CIFAR10 A2 (k=500)accuracy: 35.5 -> 45.8Improved
use GDumb instead of ARMcontinual learner · Split-CIFAR10 A3accuracy: 26.4 -> 61.3Improved
increase memory from 200 to 1K (ablation)GDumb · Split-CIFAR10 A1accuracy: 35.0 -> 61.3Improved
increase memory from 200 to 1KER-MIR · Split-CIFAR10 A1accuracy: 29.8 -> 47.6Improved
use GDumb instead of iCARL/BiC with 20 taskscontinual learner · CIFAR100 B2accuracy: 63.8 -> 45.2Hurt
use GDumb instead of UCIR (CNN) with 50 taskscontinual learner · CIFAR100 B3accuracy: 49.3 -> 58.4Improved
use GDumb (CNN) instead of PODNet (CNN)continual learner · CIFAR100 B3accuracy: 58.0 -> 58.4No clear effect
use GDumb instead of iTAMLcontinual learner · MNIST B1accuracy: 97.9 -> 97.8No clear effect
use GDumb instead of iCARL with k=4500continual learner · Split-TinyImageNet C2accuracy: 48.55 -> 45.50Hurt
use GDumb instead of iCARL with k=9000continual learner · Split-TinyImageNet C2accuracy: 49.94 -> 57.27Improved

Later papers that compared with this method

13 later accepted papers report GDumb as a baseline in their results tables; 10 of them report a better number on most of the results they share with it.

5 more not listed.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 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