List your company
ResearchFormerly WizWand

Image Classification on MNIST standard (test) (Class Disparity Metrics)

95.16Macro AccuracyBounded Adaptive
72.009678.019884.0390.0402Jun 2, 2025
Updated 4mo ago

Evaluation results

MethodLinks
Bounded Adaptiveepsilon=4.095*, DP-HPO=true2025.0695.1680.39
Constant Clippingepsilon=4.095*, DP-HPO=true2025.0695.1577.54
Unbounded Adaptiveepsilon=4.095*, DP-HPO=true2025.0694.5673.77
Constant Clippingepsilon=3.509, DP-HPO=true2025.0694.3569.9
Constant Clippingepsilon=2.916, DP-HPO=true2025.0694.1569.5
AUTOepsilon=4.095*, DP-HPO=true2025.0693.9570.27
Bounded Adaptiveepsilon=3.509, DP-HPO=true2025.0693.8873.88
Unbounded Adaptiveepsilon=3.509, DP-HPO=true2025.0693.8770.3
AUTOepsilon=3.509, DP-HPO=true2025.0693.8468.27
Bounded Adaptiveepsilon=2.916, DP-HPO=true2025.0693.6673.2
AUTOepsilon=2.916, DP-HPO=true2025.0692.8665.27
Unbounded Adaptiveepsilon=2.916, DP-HPO=true2025.0692.7266.14
Unbounded Adaptiveepsilon=2.311, DP-HPO=true2025.0692.5662.95
Constant Clippingepsilon=2.311, DP-HPO=true2025.0692.2354.5
Bounded Adaptiveepsilon=2.311, DP-HPO=true2025.0691.7760.23
AUTOepsilon=2.311, DP-HPO=true2025.0691.7460.75
AUTOepsilon=1.743, DP-HPO=true2025.0689.4352.62
Bounded Adaptiveepsilon=1.743, DP-HPO=true2025.0683.240.23
Unbounded Adaptiveepsilon=1.743, DP-HPO=true2025.0681.9742.87
Constant Clippingepsilon=1.743, DP-HPO=true2025.0672.920.11