AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Efficient Statistical Assessment of Neural Network Corruption Robustness

Karim TIT, Teddy Furon, Mathias ROUSSET

NeurIPS 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization · 2 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 Last Particle instead of DeepPolyLast Particle · MNIST eps=0.015certified (%): 82 -> 99Improved
use Last Particle instead of DeepPolyLast Particle · MNIST eps=0.03certified (%): 62 -> 97Improved
use Last Particle instead of DeepPolyLast Particle · MNIST eps=0.06certified (%): 28 -> 93Improved
use Last Particle instead of DeepPolyLast Particle · MNIST eps=0.1certified (%): 22 -> 85Improved
use Last Particle instead of DeepPolyLast Particle · MNIST eps=0.015runtime (sec.): 5.69 -> 1.04 ± 0.005Improved
use Last Particle instead of DeepPolyLast Particle · MNIST eps=0.1runtime (sec.): 8.84 -> 0.96 ± 0.02Improved
increase epsilon from 0.015 to 0.1Last Particle · MNISTruntime (sec.): 1.04 ± 0.005 -> 0.96 ± 0.02Improved
increase epsilon from 0.015 to 0.1ERAN DeepPoly · MNISTruntime (sec.): 5.69 -> 8.84Hurt
increase number of particles N from 2 to 20Last Particle · idealized linear casescore function calls: 167 -> 1470Hurt
increase epsilon from 0.02 to 0.06Last Particle · ImageNet ResNet50certified (%): 81 -> 59Hurt
increase epsilon from 0.02 to 0.06Last Particle · ImageNet MobileNetavg runtime (sec.): 20.78 ± 0.74 -> 14.5 ± 0.11Improved
use Last Particle vs ERAN on ACAS Xu propertiesLast Particle · ACAS Xunumber of cases certified: 107 certified, 9 false positivesMixed

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