WRN-28-10 with reSGHMC

Reported on 2 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision2 results

Image ClassificationonCIFAR-10
Percentage correct· 2020-08-12
97.42
best: 99.5 (ViT-H/14)
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC arXiv:2008.05367
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2020-08-12
84.38
best: 96.08 (EffNet-L2 (SAM))
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC arXiv:2008.05367