AI Paper Atlas

Research map / Continual learning / Online and replay-based

Calibrating CNNs for Lifelong Learning

Pravendra Singh, Vinay Kumar Verma, Pratik Mazumder, Lawrence Carin, Piyush Rai

NeurIPS 2020 · Continual learning › Online and replay-based · approach: Gradient projection and plasticity preservation · 35 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 spatial calibration module (ablation)CCLL<1,1> · Split CIFAR-10/100A6 accuracy: 87.85 -> 89.58Improved
add channel-wise calibration module (ablation)CCLL<1,1> · Split CIFAR-10/100A6 accuracy: 89.06 -> 89.58Improved
add both calibration modules (SCM and CCM) (ablation)CCLL base network · Split CIFAR-10/100A6 accuracy: 67.06 -> 89.58Improved
increase alpha (channels per group in SCM) from 1 to 8 (ablation)CCLL · Split CIFAR-10/100A6 accuracy: 89.58 -> 90.99Improved
train base module on first task instead of leaving it untrained (freeze random base) (ablation)CCLL · Split CIFAR-10/100accuracyImproved
initialize calibration parameters from previous task instead of from scratch (ablation)CCLL · Split CIFAR-10/100accuracyImproved
add batch normalization layers to LeNet (ablation)CCLL<1,1> · CIFAR-100 10 classes per taskaccuracy: 63.71% -> 71.52%Improved
pre-train base module on CIFAR-10 and keep frozen instead of training on first SVHN task (ablation)CCLL · SVHN (5 tasks)A5 accuracy: 98.2% -> 97.9%No clear effect
finetune CIFAR-10 pretrained base on first SVHN task (ablation)CCLL · SVHN (5 tasks)A5 accuracy: 98.2% -> 98.4%No clear effect
use CCLL instead of the closest replay-based baselineResNet-18 continual learner · ImageNet-100/10final top-5 accuracy: 74.1 -> 97.9Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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