AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

Evolutionary Neural Architecture Search for Transformer in Knowledge Tracing

Shangshang Yang, Xiaoshan Yu, Ye Tian, Xueming Yan, Haiping Ma, Xingyi Zhang

NeurIPS 2023 · Architectures and efficiency › Neural architecture search · approach: AutoML pipeline search · 12 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
ENAS-KT architecture instead of SAINT+Transformer KT · EdNetAUC: 0.7916 -> 0.8062Improved
ENAS-KT instead of NAS-CellKT · EdNetAUC: 0.7796 -> 0.8062Improved
use selected features instead of all features (concat) (ablation)Transformer KT · EdNetAUC: 0.7937 -> 0.7958Improved
hierarchical fusion instead of concat (ablation)Transformer KT · EdNetAUC: 0.7958 -> 0.7987Improved
add convolution local operations (ablation)Transformer KT · EdNetAUC: 0.7987 -> 0.8012Improved
search architecture (vs fixed Conv 1x3) (ablation)ENAS-KT · EdNetAUC: 0.8012 -> 0.8062Improved
remove searched architecture (use SAINT+ model) (ablation)ENAS-KT · EdNetAUC: 0.8062 -> 0.7987Hurt
remove hierarchical fusion (use concat) (ablation)ENAS-KT · EdNetAUC: 0.8062 -> 0.8041Hurt
remove selected features (use all) (ablation)ENAS-KT · EdNetAUC: 0.8062 -> 0.8030Hurt
increase SAINT hidden size to 192SAINT · EdNetAUC: 0.7825 -> 0.7828No clear effect
increase SAINT+ hidden size to 192SAINT+ · EdNetAUC: 0.7916 -> 0.7921No clear effect
use cheaper search (small supernet) (ablation)ENAS-KT · EdNetAUC: 0.8062 -> 0.8036Hurt

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