AI Paper Atlas

Research map / Continual learning / Online and replay-based

BayesPCN: A Continually Learnable Predictive Coding Associative Memory

Jinsoo Yoo, Frank Wood

NeurIPS 2022 · Continual learning › Online and replay-based · approach: Theoretical analysis of forgetting · 2 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 BayesPCN instead of online GPCNBayesPCN · hetero-associative Mask CIFAR10, 1024 imagesMSE: 0.0791 -> 0.0019Improved
use BayesPCN instead of online GPCNBayesPCN · hetero-associative Dropout CIFAR10, 1024 imagesMSE: 0.0073 -> 0.0001Improved
use BayesPCN instead of online GPCN at 1024 imagesBayesPCN · White Noise CIFAR10 auto-associativeMSE: 0.0210 -> 0.0337Hurt
use BayesPCN instead of online GPCN at 1024 imagesBayesPCN · White Noise Tiny ImageNet auto-associativeMSE: 0.0181 -> 0.6606Hurt
use BayesPCN instead of online GPCN at 128 imagesBayesPCN · White Noise CIFAR10 auto-associativeMSE: 0.0103 -> 0.0017Improved
add forget operation to BayesPCN at 1024 images (ablation)BayesPCN · White Noise CIFAR10MSE: 0.0337 -> 0.0188Improved
add forget operation to BayesPCN at 1024 images (ablation)BayesPCN · White Noise Tiny ImageNetMSE: 0.6606 -> 0.0176Improved
add forget operation to BayesPCN at 1024 images (ablation)BayesPCN · Mask CIFAR10MSE: 0.0019 -> 0.0465Hurt
add forget operation to BayesPCN at 1024 images (ablation)BayesPCN · Mask Tiny ImageNetMSE: 0.0000 -> 0.0235Hurt
increase forget strength beta (0 to 0.02) (ablation)BayesPCN · CIFAR10, 1024 imagesrecall MSE on recent batchesImproved
use BayesPCN instead of MHN under high query noise (dropout 75%)BayesPCN · Dropout CIFAR10, 128 imagesMSE: 0.3840 -> 0.0000Improved
use BayesPCN instead of MHN under high noise white noiseBayesPCN · White Noise CIFAR10, 128 imagesMSE: 0.0052 -> 0.0755Hurt

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