AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Constrained Optimization Based Low-Rank Approximation of Deep Neural Networks

Chong Li, C. J. Richard Shi

ECCV 2018 · Architectures and efficiency › RNNs and normalization · approach: Theory of depth and width · 60 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 COBLA instead of POS-CVX after fine-tuning (budget 0.7, Equ. 3)SqueezeNet low-rank · ImageNet valtop-1 accuracy (%): 55.7% vs baseline -2.0%Improved
use COBLA instead of POS-CVX (budget 0.5, Equ. 3)SqueezeNet low-rank · ImageNet valtop-1 accuracy (%): 52.6% vs baseline -7.4%Improved
use COBLA instead of POS-CVX (budget 0.5, Equ. 4)SqueezeNet low-rank · ImageNet valtop-5 accuracy (%): 76.2% vs baseline -4.1%Improved
use COBLA instead of POS-CVX (budget 0.6, Equ. 4)SqueezeNet low-rank · ImageNet valtop-1 accuracy (%): 54.3% vs baseline -3.8%Improved
use COBLA instead of R-VBMF under severe budget, 10 epoch fine-tuningSqueezeNet low-rank · ImageNet valtop-1 accuracy (%): 15.9% vs near chanceImproved
use Equ. 3 decomposition instead of Equ. 4 (before fine-tuning) (ablation)COBLA · ImageNet valtop-1 accuracy (%): COBLA much better with Equ. 3 before fine-tuningImproved
switch decomposition scheme from Equ. 4 to Equ. 3 after fine-tuning (ablation)COBLA · ImageNet valtop-1 accuracy (%): difference within 1%No clear effect
use sub-sampled gradient (5% of data) instead of exact gradient in SQP (ablation)COBLA · CIFAR-10 NINtop-1 accuracy (%): within 1%No clear effect
use COBLA allocation instead of manual trial-and-error ranks [27]VGG-16 low-rank · ImageNet valtop-5 accuracy (%): 89.8% -> 89.8% with 12% less costImproved
push COBLA to 30% cost reduction versus [27]VGG-16 low-rank · ImageNet valtop-5 accuracy (%): 89.8% -> 88.9%Hurt
add data-dependent decomposition (ablation)low-rank approximation · ImageNet valaccuracyNo 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