AI Paper Atlas

Research map / Continual learning / Online and replay-based

CLeAR: Continual Learning on Algorithmic Reasoning for Human-like Intelligence

Bong Gyun Kang, HyunGi Kim, Dahuin Jung, Sungroh Yoon

NeurIPS 2023 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 0 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 CLeAR to sequential training (ablation)RNN · Regular CN-MA-PC-EPCL Final accuracy (%): 47.50 -> 96.17Improved
add CLeAR to sequential training (ablation)LSTM · Regular CN-MA-PC-EPCL Final accuracy (%): 50.65 -> 96.49Improved
add CLeAR to sequential training (ablation)RNN · Regular CN-MA-PC-EPBWT: -69.80 -> -0.37Improved
add CLeAR to sequential training (ablation)Tape-RNN · Regular EP-PC-MA-CNCL Final accuracy (%): 47.56 -> 74.45Improved
add CLeAR to sequential training (ablation)Stack-RNN · Regular EP-PC-MA-CNBWT: -59.82 -> 1.45Improved
add CLeAR to sequential training (ablation)Stack-RNN · Regular CN-MA-PC-EPBWT: -69.04 -> -10.40Improved
change task order from CN-MA-PC-EP to EP-PC-MA-CN with CLeARRNN · Regular in-hierarchyCL Final accuracy (%): 96.17 -> 99.80Improved
change task order from CN-MA-PC-EP to EP-PC-MA-CN with CLeARStack-RNN · Regular in-hierarchyCL Final accuracy (%): 87.09 -> 99.28Improved
continual learning with CLeAR instead of joint trainingLSTM · High-correlation Modular Arithmeticaccuracy (%): 93.07 -> 96.25Improved
continual learning with CLeAR instead of joint trainingStack-RNN · High-correlation Bucket Sortaccuracy (%): 93.49 -> 92.44No clear effect
continual learning with CLeAR instead of joint trainingTape-RNN · In-hierarchy Regular EP-PC-MA-CNaccuracy (%): 59.61 -> 74.45Improved
continual learning with CLeAR instead of joint trainingStack-RNN · In-hierarchy Context Free DT-RS-SM-COaccuracy (%): 69.64 -> 69.95No 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