# Image Classification on MNIST (test) (Efficiency/Overhead Metrics)

Source: [Image Classification on MNIST (test) (Efficiency/Overhead Metrics)](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-efficiency-overhead-metrics>)

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

Data updated: 2026-03-10T13:39:14.030Z

Evaluation of Image Classification models on the MNIST test set, focusing on performance gains related to system efficiency, specifically measuring Delay Gain and Overhead Gain.

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) | MNIST (test) |
| Metric | Delay Gain |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Delay Gain](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-efficiency-overhead-metrics.md?metric=Delay%20Gain>): Higher is better; unit: PERCENT.

- [Overhead Gain](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-efficiency-overhead-metrics.md?metric=Overhead%20Gain>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Delay Gain | Result date | Report paper |
| --- | --- | --- | --- |
| AA HSFL-ll | 22 | Not recorded | 69afa5b12d603e6de3e43dd1 |
| AA HSFL-ll | 20 | Not recorded | 69afa5b12d603e6de3e43dd1 |
| AA HSFL-ll | 7.25 | Not recorded | 69afa5b12d603e6de3e43dd1 |

### AA HSFL-ll

Delay Gain: 22. Result ID: 69b01eea6dbb7075b95a46b6.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b12d603e6de3e43dd1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69afa5b12d603e6de3e43dd1.md>)

Recorded evaluation configuration: {"Model":"AlexNet","Acc. Target(%)":"90"}. This record does not assert independent reproduction.

### AA HSFL-ll

Delay Gain: 20. Result ID: 69b01eea6dbb7075b95a46b7.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b12d603e6de3e43dd1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69afa5b12d603e6de3e43dd1.md>)

Recorded evaluation configuration: {"Model":"AlexNet","Acc. Target(%)":"94"}. This record does not assert independent reproduction.

### AA HSFL-ll

Delay Gain: 7.25. Result ID: 69b01eea6dbb7075b95a46b5.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b12d603e6de3e43dd1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69afa5b12d603e6de3e43dd1.md>)

Recorded evaluation configuration: {"Model":"AlexNet","Acc. Target(%)":"85"}. This record does not assert independent reproduction.

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