AI Paper Atlas

Research map / Continual learning / Domain and prompt-based

A Unified Approach to Domain Incremental Learning with Memory: Theory and Algorithm

Haizhou Shi, Hao Wang

NeurIPS 2023 · Continual learning › Domain and prompt-based · approach: Domain-incremental learning with memory · 10 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 fixed replay coefficients (DER++) with adaptive learned coefficients and domain alignment (UDIL)DER++ · R-MNISTaverage accuracy: 84.258 -> 86.635Improved
adaptive coefficients vs DER++ on R-MNIST forgettingDER++ · R-MNISTforgetting: 13.692 -> 8.506Improved
adaptive coefficients vs best baseline on Seq-CORe50DER++ · Seq-CORe50average accuracy: 78.629 -> 82.103Improved
adaptive coefficients vs DER++ on Seq-CORe50 forward transferDER++ · Seq-CORe50forward transfer: 22.488 -> 31.215Improved
adaptive coefficients vs CLS-ER on HD-BallsCLS-ER · HD-Ballsaverage accuracy: 85.844 -> 86.872Improved
UDIL vs DER++ on permutation MNIST accuracyDER++ · P-MNISTaverage accuracy: 92.950 -> 92.666Hurt
UDIL vs DER++ on permutation MNIST forgettingDER++ · P-MNISTforgetting: 3.378 -> 2.853Improved
add memory replay buffer (ER) to fine-tuningFine-tune · Seq-CORe50average accuracy: 31.832 -> 66.605Improved

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