# Dimensionality Reduction on CIFAR-10

Source: [Dimensionality Reduction on CIFAR-10](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10>)

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

Data updated: 2026-02-26T05:47:45.186Z

Evaluation of dimensionality reduction techniques applied to the CIFAR-10 dataset, typically assessed using metrics like KL divergence, Trust scores, 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) | CIFAR-10 |
| 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-cifar-10.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-cifar-10.md?metric=KL%20Divergence%20(0.1)>): Lower is better; unit: SCALAR.

- [KL Divergence (1)](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10.md?metric=KL%20Divergence%20(1)>): Lower is better; unit: SCALAR.

- [MRRE Score](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10.md?metric=MRRE%20Score>): Higher is better; unit: SCALAR.

- [Continuity Score](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10.md?metric=Continuity%20Score>): Higher is better; unit: SCALAR.

- [Trust Score](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10.md?metric=Trust%20Score>): Higher is better; unit: SCALAR.

- [RMSE](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10.md?metric=RMSE>): Lower is better; unit: SCALAR.

- [Data MSE](<https://www.sota2.com/research/sota/dimensionality-reduction-on-cifar-10.md?metric=Data%20MSE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | KL Divergence (0.01) | Result date | Report paper |
| --- | --- | --- | --- |
| TopoAE | 0.556 | Not recorded | 68fa58241e209131ce9abb73 |
| PCA | 0.591 | Not recorded | 68fa58241e209131ce9abb73 |
| UMAP | 0.617 | Not recorded | 68fa58241e209131ce9abb73 |
| TSNE | 0.627 | Not recorded | 68fa58241e209131ce9abb73 |
| AE | 0.668 | Not recorded | 68fa58241e209131ce9abb73 |

### TopoAE

KL Divergence (0.01): 0.556. Result ID: 698ba2457a967eddb770340e.

[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.

### PCA

KL Divergence (0.01): 0.591. Result ID: 698ba2457a967eddb770340a.

[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): 0.617. Result ID: 698ba2457a967eddb770340c.

[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.

### TSNE

KL Divergence (0.01): 0.627. Result ID: 698ba2457a967eddb770340b.

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

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

### AE

KL Divergence (0.01): 0.668. Result ID: 698ba2457a967eddb770340d.

[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.
