# Dynamic Feature Selection on Image datasets (held-out feature subsets)

Source: [Dynamic Feature Selection on Image datasets (held-out feature subsets)](<https://www.sota2.com/research/sota/dynamic-feature-selection-on-image-datasets-held-out-feature-subsets>)

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

Data updated: 2026-05-13T10:56:00.332Z

Evaluation of Dynamic Feature Selection performance on a specific collection of held-out feature subsets derived from various image datasets, measured using Avg AUAC and Avg Rank.

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 | Dynamic Feature Selection |
| Dataset (including any stated split) | Image datasets (held-out feature subsets) |
| Metric | Avg AUAC |
| Unit | SCALAR |
| 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

- [Avg AUAC](<https://www.sota2.com/research/sota/dynamic-feature-selection-on-image-datasets-held-out-feature-subsets.md?metric=Avg%20AUAC>): Higher is better; unit: SCALAR.

- [Avg Rank](<https://www.sota2.com/research/sota/dynamic-feature-selection-on-image-datasets-held-out-feature-subsets.md?metric=Avg%20Rank>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Avg AUAC | Result date | Report paper |
| --- | --- | --- | --- |
| Hyper-DFS | 87.57 | Not recorded | 6a0405e0dbc836b87d8278e6 |
| VIP | 85.02 | Not recorded | 6a0405e0dbc836b87d8278e6 |
| MoE | 83.94 | Not recorded | 6a0405e0dbc836b87d8278e6 |

### Hyper-DFS

Avg AUAC: 87.57. Result ID: 6a0458a547f34ce924eb09a0.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405e0dbc836b87d8278e6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0405e0dbc836b87d8278e6.md>)

Recorded evaluation configuration: {"evaluation mode":"zero-shot"}. This record does not assert independent reproduction.

### VIP

Avg AUAC: 85.02. Result ID: 6a0458a547f34ce924eb099e.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405e0dbc836b87d8278e6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ffa1e209131cea484ad.md>)

Recorded evaluation configuration: {"evaluation mode":"zero-shot"}. This record does not assert independent reproduction.

### MoE

Avg AUAC: 83.94. Result ID: 6a0458a547f34ce924eb099f.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405e0dbc836b87d8278e6.md>)

Recorded evaluation configuration: {"evaluation mode":"zero-shot"}. This record does not assert independent reproduction.

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