AI Paper Atlas

Research map / Continual learning / Class-incremental

Learning to Learn: How to Continuously Teach Humans and Machines

Parantak Singh, You Li, Ankur Sikarwar, Stan Weixian Lei, Difei Gao, Morgan B. Talbot, Ying Sun, Mike Zheng Shou, Gabriel Kreiman, Mengmi Zhang

ICCV 2023 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 6 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
choose a different class order (curriculum) in online class-incremental learningVanilla CL · MNIST/FashionMNIST/CIFAR10average accuracy alpha: alpha ranged from 19% to 26% depending on curriculumImproved
use CD-ranked curricula instead of random rankingCurriculum Designer · MNIST and CIFAR10 (Recall@K)Recall@K: peaks of 0.5, 0.2, 1 at K=2, 5, 10 for MNIST, FashionMNIST, CIFAR10Improved
use CD instead of random designerCurriculum Designer · FashionMNISTRecall@KHurt
use Euclidean instead of cosine distance for class prototypes (ablation)Curriculum Designer · MNISTRecall@KNo clear effect
use OTDD instead of cosine distance (ablation)Curriculum Designer · MNISTRecall@KHurt
use earlier layers (6 or 11) rather than layer 12 features (ablation)Curriculum Designer · MNISTRecall@KHurt
replace SqueezeNet with ResNet34/ResNet18 teacher (ablation)Curriculum Designer · MNISTRecall@KMixed
pre-train the teacher on MNIST instead of ImageNet-100 (ablation)Curriculum Designer · MNISTRecall@KNo clear effect
use a pre-trained teacher rather than randomly initialized teacher (ablation)Curriculum Designer · MNISTRecall@KImproved
compare CL algorithm rankings between algorithms versus against random designercurricula ranking agreement · MNIST (paradigm-I)Spearman correlation: 0.0002 -> 0.26 (algorithm-random vs between-algorithms)Improved
use curricula ranked by algorithms for human learners versus randomNOD human experiment · NODcurriculum discrepancy H: H decrease of 0.13 from random-humans to algorithm-humansImproved
apply CD curricula to humansCurriculum Designer · NODcurriculum discrepancy HNo clear effect

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