# Image Classification on CIFAR-100 (Weight Norm and AutoAtt Metrics)

Source: [Image Classification on CIFAR-100 (Weight Norm and AutoAtt Metrics)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-weight-norm-and-autoatt-metrics>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:48:02.652Z

Evaluation of Image Classification performance on the CIFAR-100 dataset, reporting metrics related to weight norms (\|\|W\|\|2), natural accuracy (Nat Acc), attention (AutoAtt), and the specific regularization loss term.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Image Classification |
| Dataset (including any stated split) | CIFAR-100 |
| Metric | L2 Norm (\|\|W\|\|2) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [L2 Norm (\|\|W\|\|2)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-weight-norm-and-autoatt-metrics.md?metric=L2%20Norm%20(%7C%7CW%7C%7C2)>): Lower is better; unit: SCALAR.

- [Nat Acc](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-weight-norm-and-autoatt-metrics.md?metric=Nat%20Acc>): Higher is better; unit: PERCENT.

- [AutoAtt](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-weight-norm-and-autoatt-metrics.md?metric=AutoAtt>): Higher is better; unit: SCALAR.

- [Reg Loss (Xent + wd\*\|\|W\|\|^2/2)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-weight-norm-and-autoatt-metrics.md?metric=Reg%20Loss%20(Xent%20%2B%20wd*%7C%7CW%7C%7C%5E2%2F2)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | L2 Norm (\|\|W\|\|2) | Result date | Report paper |
| --- | --- | --- | --- |
| Adaptive Weight Decay (AWD) | 13.41 | Not recorded | 68fa71551e209131cea508f2 |
| Non-adaptive Weight Decay | 51.32 | Not recorded | 68fa71551e209131cea508f2 |

### Adaptive Weight Decay (AWD)

L2 Norm (\|\|W\|\|2): 13.41. Result ID: 6991c5973a6c3148490a7e5e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71551e209131cea508f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71551e209131cea508f2.md>)

Recorded evaluation configuration: {"lambda\_awd":"0.022","Optimizer":"SGD"}. This record does not assert independent reproduction.

### Non-adaptive Weight Decay

L2 Norm (\|\|W\|\|2): 51.32. Result ID: 6991c5973a6c3148490a7e5d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71551e209131cea508f2.md>)

Recorded evaluation configuration: {"lambda\_wd":"0.0005","Optimizer":"SGD"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.