AI Paper Atlas

Research map / Continual learning / Online and replay-based

Continual Learning in Low-rank Orthogonal Subspaces

Arslan Chaudhry, Naeemullah Khan, Puneet Dokania, Philip Torr

NeurIPS 2020 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 32 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 orthogonal task subspaces without episodic memoryorthog-subspace on Permuted MNIST · Permuted MNISTAccuracy: 79.5 -> 86.6Improved
Use orthogonal task subspaces without episodic memoryorthog-subspace on Permuted MNIST · Permuted MNISTForgetting: 0.12 -> 0.04Improved
Use orthogonal task subspaces without episodic memoryorthog-subspace on Rotated MNIST · Rotated MNISTAccuracy: 70.9 -> 80.1Improved
Use orthogonal task subspaces without episodic memoryorthog-subspace on Rotated MNIST · Rotated MNISTForgetting: 0.24 -> 0.14Improved
Use orthogonal subspaces, replay memory, and Stiefel-constrained weightsorthog-subspace on Split CIFAR · Split CIFARAccuracy: 59.6 -> 64.3Improved
Use orthogonal subspaces, replay memory, and Stiefel-constrained weightsorthog-subspace on Split miniImageNet · Split miniImageNetAccuracy: 49.8 -> 51.4Improved
Add projection to ER-Ring without Stiefel constraint (ablation)orthog-subspace on Split CIFAR · Split CIFARAccuracy: 59.6 -> 61.2Improved
Add projection to ER-Ring without Stiefel constraint (ablation)orthog-subspace on Split miniImageNet · Split miniImageNetAccuracy: 49.8 -> 49.5No clear effect
Add the Stiefel constraint to projection plus ER (ablation)orthog-subspace on Split CIFAR · Split CIFARAccuracy: 61.2 -> 64.3Improved
Add the Stiefel constraint to projection plus ER (ablation)orthog-subspace on Split miniImageNet · Split miniImageNetAccuracy: 49.5 -> 51.4Improved
Increase replay memory from 2 to 5 examples per class per task (ablation)orthog-subspace on Split CIFAR · Split CIFARAccuracy: 64.7 -> 67.3Improved
Increase replay memory from 2 to 5 examples per class per task (ablation)orthog-subspace on Split miniImageNet · Split miniImageNetAccuracy: 53.4 -> 58.2Improved

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