# Image Classification on MNIST 5000 train, 8000 (test)

Source: [Image Classification on MNIST 5000 train, 8000 (test)](<https://www.sota2.com/research/sota/image-classification-on-mnist-5000-train-8000-test>)

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

Data updated: 2026-02-26T05:28:58.506Z

Evaluation of Image Classification performance, including accuracy and calibration metrics (NLL, Brier, ECE, MCE), on a specific split of the MNIST dataset (5000 training, 8000 testing).

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 5000 train, 8000 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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](<https://www.sota2.com/research/sota/image-classification-on-mnist-5000-train-8000-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [NLL](<https://www.sota2.com/research/sota/image-classification-on-mnist-5000-train-8000-test.md?metric=NLL>): Lower is better; unit: SCALAR.

- [Brier](<https://www.sota2.com/research/sota/image-classification-on-mnist-5000-train-8000-test.md?metric=Brier>): Lower is better; unit: SCALAR.

- [ECE](<https://www.sota2.com/research/sota/image-classification-on-mnist-5000-train-8000-test.md?metric=ECE>): Lower is better; unit: PERCENT.

- [MCE](<https://www.sota2.com/research/sota/image-classification-on-mnist-5000-train-8000-test.md?metric=MCE>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| RIEM-LA | 96.74 | Not recorded | 68fa6faf1e209131cea46e23 |
| RIEM-LA (BATCHES) | 95.67 | Not recorded | 68fa6faf1e209131cea46e23 |
| LIN-RIEM-LA | 95.44 | Not recorded | 68fa6faf1e209131cea46e23 |
| LIN-RIEM-LA (BATCHES) | 95.14 | Not recorded | 68fa6faf1e209131cea46e23 |
| MAP | 95.02 | Not recorded | 68fa6faf1e209131cea46e23 |
| LIN-LA | 94.91 | Not recorded | 68fa6faf1e209131cea46e23 |
| VANILLA LA | 88.69 | Not recorded | 68fa6faf1e209131cea46e23 |

### RIEM-LA

Accuracy: 96.74. Result ID: 6990e6d63a6c314849042520.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15","Batching Strategy":"Full"}. This record does not assert independent reproduction.

### RIEM-LA (BATCHES)

Accuracy: 95.67. Result ID: 6990e6d63a6c314849042521.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15","Batch size":"1000"}. This record does not assert independent reproduction.

### LIN-RIEM-LA

Accuracy: 95.44. Result ID: 6990e6d63a6c314849042522.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15","Batching Strategy":"Full"}. This record does not assert independent reproduction.

### LIN-RIEM-LA (BATCHES)

Accuracy: 95.14. Result ID: 6990e6d63a6c314849042523.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15","Batch size":"1000"}. This record does not assert independent reproduction.

### MAP

Accuracy: 95.02. Result ID: 6990e6d63a6c31484904251d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15"}. This record does not assert independent reproduction.

### LIN-LA

Accuracy: 94.91. Result ID: 6990e6d63a6c31484904251f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69541e209131cea2192f.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15"}. This record does not assert independent reproduction.

### VANILLA LA

Accuracy: 88.69. Result ID: 6990e6d63a6c31484904251e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6faf1e209131cea46e23.md>)

Recorded evaluation configuration: {"Prior Precision":"Optimized","Monte Carlo samples":"25","Backbone":"CNN","Bin count (M)":"15"}. This record does not assert independent reproduction.

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