AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

Harmless Overfitting: Using Denoising Autoencoders in Estimation of Distribution Algorithms

Malte Probst, Franz Rothlauf

JMLR 2020 · Probabilistic and scientific modeling › Bayesian optimization · approach: Multi-fidelity hyperparameter optimization · 11 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 DAE-EDA instead of BOA on 5-Traps 100 bitEDA model · 5-Traps 100 bitunique evaluations: 190,011 -> 62,599Improved
use DAE-EDA instead of BOA on onemaxEDA model · ONEMAX100unique evaluations: 6,259 -> 7,005Hurt
use DAE-EDA instead of BOA on NK landscapesEDA model · NK landscapesunique evaluations: 32,015 -> 17,472 (NK n=30,k=4,i=1); 67,939 -> 189,613 (NK n=30,k=4,i=2)Mixed
use DAE-EDA instead of BOA on HIFFEDA model · HIFFunique evaluations: 11,991 -> 22,066 (HIFF64); 51,008 -> 50,447 (HIFF128)Mixed
use DAE-EDA instead of BOAEDA model · All problem instancesCPU time (sec): 205,244 -> 2,633 (4-Traps 120 bit)Improved
use DAE-EDA sampling instead of DAE-guided local searchHIFF 256 bit · HIFF 256 bitaverage evaluations: 1,355,500 -> 220,650Improved
lower DAE corruption noise c (weaker regularization) (ablation)DAE-EDA · 50 bit 5-Traps, N=200average generations to best solution: 22.0 -> 30.0 generations (c=0.2 to c=0.05)Improved
lower DAE corruption noise c (weaker regularization) (ablation)DAE-EDA · 50 bit 5-Traps, N=200fitness of best sample (avg best): 43.68 -> 46.52Improved
lower DAE corruption noise c (weaker regularization) (ablation)DAE-EDA · 64 bit HIFF, N=200fitness of best sample (avg best): 305.10 -> 346.80Improved
lower DAE corruption noise c (weaker regularization) (ablation)DAE-EDA · NK n=30, k=4, i=1, N=200fitness of best sample (avg best): 22.84 -> 22.92No clear effect
use DAE-EDA with c=0.05 instead of BOAEDA model · 50 bit 5-Traps, N=200average generations to best solution: 18.7 -> 30.0Improved
lower DAE corruption noise c (ablation)DAE-EDA · 50 bit 5-Trapsidentity function degree tauImproved

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