AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

DCP–NAS: Discrepant Child–Parent Neural Architecture Search for 1-bit CNNs

Yanjing Li, Sheng Hua Xu, Xianbin Cao, Li’an Zhuo, Baochang Zhang, Tian Wang, Guodong Guo

IJCV 2023 · Architectures and efficiency › Neural architecture search · approach: Efficient NAS search strategies · 3 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 tangent direction constraint (ablation)DCP-NAS-L · ImageNettop-1 accuracy: 66.7 -> 68.3Improved
add reconstruction error term (ablation)DCP-NAS-L · ImageNettop-1 accuracy: 66.7 -> 68.2Improved
add both tangent constraint and reconstruction error (ablation)DCP-NAS-L · ImageNettop-1 accuracy: 66.7 -> 72.4Improved
add GGN approximation of the Hessian (ablation)DCP-NAS-L · ImageNetsearch cost (GPU days): 27.9 -> 2.9Improved
add decoupled optimization (ablation)DCP-NAS-L (tangent propagation + GGN) · ImageNettop-1 accuracy: 71.2 -> 72.4Improved
add decoupled optimization (ablation)DARTS · ImageNettop-1 accuracy: 73.1 -> 73.8Improved
add decoupled optimization (ablation)PC-DARTS · ImageNettop-1 accuracy: 75.8 -> 76.6Improved
use F-norm instead of cosine similarity or L1-norm for tangent distance (ablation)DCP-NAS-S · ImageNettop-1 accuracy: 62.5 -> 63.0Improved
keep pooling operations in search space (ablation)DCP-NAS-S · ImageNettop-1 accuracy: 59.1 -> 63.0Improved
keep skip connections in search space (ablation)DCP-NAS-S · ImageNettop-1 accuracy: 60.0 -> 63.0Improved
keep activations of first depth-wise conv real-valued instead of binarizing (ablation)DCP-NAS-L · ImageNettop-1 accuracy: 70.1 -> 72.4Improved
keep preprocessing activations real-valued instead of binarizing (ablation)DCP-NAS-S · ImageNettop-1 accuracy: 60.7 -> 63.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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