AI Paper Atlas

Research map / Continual learning / Class-incremental

Loss Decoupling for Task-Agnostic Continual Learning

Yan-Shuo Liang, Wu-Jun Li

NeurIPS 2023 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 3 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 LODE to DER++DER++ · Seq-CIFAR10 (buffer 500)accuracy: 73.29 -> 75.45Improved
add LODE to DER++DER++ · Seq-CIFAR100 (buffer 500)accuracy: 42.08 -> 46.31Improved
add LODE to DER++DER++ · Seq-TinyImageNet (buffer 500)accuracy: 19.28 -> 21.15Improved
add LODE to ESMERESMER · Seq-CIFAR100 (buffer 500)accuracy: 52.37 -> 55.06Improved
add LODE to ERER · Seq-CIFAR100 (buffer 500)accuracy: 27.69 -> 41.52Improved
add LODE to ER-ACE-like ER, buffer 5120ER · Seq-CIFAR10 (buffer 5120)accuracy: 83.64 -> 83.73No clear effect
set beta1=beta2=C (remove decoupling) (ablation)LODE (DER++) · Seq-CIFAR100accuracy: 46.31 -> 42.08Hurt
set beta1=beta2=rho|Cn|/|Co| (remove decoupling) (ablation)LODE (DER++) · Seq-CIFAR100accuracy: 46.31 -> 37.49Hurt
exchange beta1 and beta2 (ablation)LODE (DER++) · Seq-CIFAR100accuracy: 46.31 -> 40.79Hurt
set beta1=beta2=C (remove decoupling) (ablation)LODE (ESMER) · Seq-CIFAR100accuracy: 55.06 -> 52.37Hurt
set beta1=beta2=rho|Cn|/|Co| (ablation)LODE (ESMER) · Seq-CIFAR100accuracy: 55.06 -> 45.64Hurt
set beta1=beta2=C on Seq-CIFAR10 (ablation)LODE (ESMER) · Seq-CIFAR10accuracy: 74.53 -> 73.08Hurt

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