AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

A Bayesian Optimization Framework for Neural Network Compression

Xingchen Ma, Amal Rannen Triki, Maxim Berman, Christos Sagonas, Jacques Cali, Matthew B. Blaschko

ICCV 2019 · Probabilistic and scientific modeling › Bayesian optimization · approach: Multi-fidelity hyperparameter optimization · 23 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 the PAC-style acquisition function with Gaussian-noise uncertainty instead of random searchBO compression of ResNet18 · ResNet18, gamma=1.0top-1 accuracy (%): 58.2 -> 62.9Improved
use the proposed BO instead of Fabolas for parameter selectionBO compression of ResNet18 · ResNet18, gamma=1.0time to converge (s): 46876 -> 1378Improved
use the proposed BO instead of FabolasBO compression of ResNet18 · ResNet18, gamma=1.0top-1 accuracy (%): 61.8 -> 62.9Improved
use knowledge distillation (function norm) fidelity instead of top-1 error as objective (ablation)FC3 SVD compression · MNIST FC3, timetime (s): 365.53 -> 79.54Improved
use knowledge distillation fidelity instead of risk on labeled data (ablation)FC3 SVD compression · MNIST FC3top-1 accuracy (%): 97.73 -> 97.78No clear effect
fine-tune the tensor-decomposed VGG-16 for 5 epochsVGG-16 compression · ILSVRC2012 valtop-1 accuracy (%): 59.86% to 69.11%Improved
fine-tune the tensor-decomposed VGG-16 for 5 epochsVGG-16 compression · ILSVRC2012 valtop-5 accuracy (%): 83.01% to 88.69%Improved
fine-tune after the second (SVD) compression stageVGG-16 compression · ILSVRC2012 valtop-1 accuracy (%): 65.41% to 68.69%Improved
increase gamma from 0.00006 to 0.0005 (ablation)ResNet18 low-rank compression · ILSVRC2012 val, no fine-tuningtop-1 accuracy (%): 36.81 -> 68.16Improved
increase gamma from 0.00006 to 0.0005 (ablation)ResNet50 low-rank compression · ILSVRC2012 val, no fine-tuningtop-1 accuracy (%): 48.93 -> 74.83Improved
increase gamma from 0.00006 to 0.0005 (ablation)ResNet50 low-rank compression · ILSVRC2012 valcompression ratio (%): 17.96 -> 39.55Mixed
increase gamma beyond 0.0005 (ablation)ResNet18/ResNet50 compression · ILSVRC2012 valtop-1 accuracy (%)No 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