# Dimensionality Reduction on Fashion-MNIST

Source: [Dimensionality Reduction on Fashion-MNIST](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist>)

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

Data updated: 2026-02-26T05:47:45.154Z

Evaluation of dimensionality reduction techniques on the Fashion-MNIST dataset, typically assessed using metrics like KL divergence, Trust, and reconstruction error (RMSE/MSE).

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 | Dimensionality Reduction |
| Dataset (including any stated split) | Fashion-MNIST |
| Metric | KL Divergence (0.01) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 5 of 5 |
| 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

- [KL Divergence (0.01)](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=KL%20Divergence%20(0.01)>): Lower is better; unit: SCALAR.

- [KL Divergence (0.1)](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=KL%20Divergence%20(0.1)>): Lower is better; unit: SCALAR.

- [KL Divergence (1)](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=KL%20Divergence%20(1)>): Lower is better; unit: SCALAR.

- [1-MRRE](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=1-MRRE>): Lower is better; unit: SCALAR.

- [Continuity Score](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=Continuity%20Score>): Higher is better; unit: SCALAR.

- [Trust Score](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=Trust%20Score>): Higher is better; unit: SCALAR.

- [RMSE](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=RMSE>): Lower is better; unit: SCALAR.

- [MSE (Data)](<https://www.sota2.com/research/sota/dimensionality-reduction-on-fashion-mnist.md?metric=MSE%20(Data)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | KL Divergence (0.01) | Result date | Report paper |
| --- | --- | --- | --- |
| PCA | 0.356 | Not recorded | 68fa58241e209131ce9abb73 |
| TopoAE | 0.392 | Not recorded | 68fa58241e209131ce9abb73 |
| TSNE | 0.405 | Not recorded | 68fa58241e209131ce9abb73 |
| UMAP | 0.424 | Not recorded | 68fa58241e209131ce9abb73 |
| AE | 0.478 | Not recorded | 68fa58241e209131ce9abb73 |

### PCA

KL Divergence (0.01): 0.356. Result ID: 698ba2457a967eddb7703400.

[Reporting paper](<https://www.sota2.com/research/paper/68fa58241e209131ce9abb73.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TopoAE

KL Divergence (0.01): 0.392. Result ID: 698ba2457a967eddb7703404.

[Reporting paper](<https://www.sota2.com/research/paper/68fa58241e209131ce9abb73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa58241e209131ce9abb73.md>)

[Implementation](<https://github.com/BorgwardtLab/topological-autoencoders>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TSNE

KL Divergence (0.01): 0.405. Result ID: 698ba2457a967eddb7703401.

[Reporting paper](<https://www.sota2.com/research/paper/68fa58241e209131ce9abb73.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### UMAP

KL Divergence (0.01): 0.424. Result ID: 698ba2457a967eddb7703402.

[Reporting paper](<https://www.sota2.com/research/paper/68fa58241e209131ce9abb73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68501e209131cea1b5d2.md>)

[Implementation](<https://github.com/lmcinnes/umap>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### AE

KL Divergence (0.01): 0.478. Result ID: 698ba2457a967eddb7703403.

[Reporting paper](<https://www.sota2.com/research/paper/68fa58241e209131ce9abb73.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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