# Image Classification on MNIST i.i.d. partitioning (test)

Source: [Image Classification on MNIST i.i.d. partitioning (test)](<https://www.sota2.com/research/sota/image-classification-on-mnist-i-i-d-partitioning-test>)

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

Data updated: 2026-04-16T07:03:06.141Z

Evaluation of Image Classification performance on the MNIST i.i.d. test set, reporting results across Complete, Regular, Ring, and Erdős–Rényi partitioning structures.

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 i.i.d. partitioning (test) |
| Metric | Accuracy (Complete) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Accuracy (Complete)](<https://www.sota2.com/research/sota/image-classification-on-mnist-i-i-d-partitioning-test.md?metric=Accuracy%20(Complete)>): Higher is better; unit: PERCENT.

- [Accuracy (Regular)](<https://www.sota2.com/research/sota/image-classification-on-mnist-i-i-d-partitioning-test.md?metric=Accuracy%20(Regular)>): Higher is better; unit: PERCENT.

- [Accuracy (Ring)](<https://www.sota2.com/research/sota/image-classification-on-mnist-i-i-d-partitioning-test.md?metric=Accuracy%20(Ring)>): Higher is better; unit: PERCENT.

- [Accuracy (Erdős–Rényi)](<https://www.sota2.com/research/sota/image-classification-on-mnist-i-i-d-partitioning-test.md?metric=Accuracy%20(Erd%C5%91s%E2%80%93R%C3%A9nyi)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Complete) | Result date | Report paper |
| --- | --- | --- | --- |
| A single SGD, no Pac-Man | 97.49 | Not recorded | 68fa7aaf1e209131cea849d0 |
| AC | 97.32 | Not recorded | 68fa7aaf1e209131cea849d0 |
| DeCaFork | 97.25 | Not recorded | 68fa7aaf1e209131cea849d0 |
| A single SGD, no self-duplication | 71.04 | Not recorded | 68fa7aaf1e209131cea849d0 |

### A single SGD, no Pac-Man

Accuracy (Complete): 97.49. Result ID: 69e0881e47f34ce924c94333.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7aaf1e209131cea849d0.md>)

Recorded evaluation configuration: {"Pac-Man node":"Absent","Self-duplication":"N/A","Optimization algorithm":"SGD"}. This record does not assert independent reproduction.

### AC

Accuracy (Complete): 97.32. Result ID: 69e0881e47f34ce924c94334.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7aaf1e209131cea849d0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7aaf1e209131cea849d0.md>)

Recorded evaluation configuration: {"Pac-Man node":"Present","Self-duplication":"True","Optimization algorithm":"Average Crossing (AC)"}. This record does not assert independent reproduction.

### DeCaFork

Accuracy (Complete): 97.25. Result ID: 69e0881e47f34ce924c94335.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7aaf1e209131cea849d0.md>)

Recorded evaluation configuration: {"Pac-Man node":"Present","Self-duplication":"True","Optimization algorithm":"DeCaFork"}. This record does not assert independent reproduction.

### A single SGD, no self-duplication

Accuracy (Complete): 71.04. Result ID: 69e0881e47f34ce924c94336.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7aaf1e209131cea849d0.md>)

Recorded evaluation configuration: {"Pac-Man node":"Present","Self-duplication":"False","Optimization algorithm":"SGD"}. This record does not assert independent reproduction.

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