# Bandit Recommendation on MovieLens 1M (top-1500 subset)

Source: [Bandit Recommendation on MovieLens 1M (top-1500 subset)](<https://www.sota2.com/research/sota/bandit-recommendation-on-movielens-1m-top-1500-subset>)

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

Data updated: 2026-06-29T07:10:42.776Z

Evaluation of bandit recommendation algorithms using regret minimization metrics on the subset of MovieLens 1 Million containing the top 1500 most-rated items.

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 | Bandit Recommendation |
| Dataset (including any stated split) | MovieLens 1M top-1500 most-rated items |
| Metric | Cumulative Regret R(T) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [Cumulative Regret R(T)](<https://www.sota2.com/research/sota/bandit-recommendation-on-movielens-1m-top-1500-subset.md?metric=Cumulative%20Regret%20R(T)>): Lower is better; unit: SCALAR.

- [Standard Error of Mean (SEM)](<https://www.sota2.com/research/sota/bandit-recommendation-on-movielens-1m-top-1500-subset.md?metric=Standard%20Error%20of%20Mean%20(SEM)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Cumulative Regret R(T) | Result date | Report paper |
| --- | --- | --- | --- |
| PCA+LinUCB | 186.7 | Not recorded | 6a41fbe6206542251d7cc9e4 |
| GraphDR+LinUCB | 273.3 | Not recorded | 6a41fbe6206542251d7cc9e4 |
| LinUCB-full | 592.6 | Not recorded | 6a41fbe6206542251d7cc9e4 |
| GraphDR-shuffled | 650.2 | Not recorded | 6a41fbe6206542251d7cc9e4 |
| JL+LinUCB | 660 | Not recorded | 6a41fbe6206542251d7cc9e4 |
| Random | 677.6 | Not recorded | 6a41fbe6206542251d7cc9e4 |

### PCA+LinUCB

Cumulative Regret R(T): 186.7. Result ID: 6a421a69fbc373592c870d64.

[Reporting paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.md>)

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

### GraphDR+LinUCB

Cumulative Regret R(T): 273.3. Result ID: 6a421a69fbc373592c870d63.

[Reporting paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.md>)

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

### LinUCB-full

Cumulative Regret R(T): 592.6. Result ID: 6a421a69fbc373592c870d62.

[Reporting paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.md>)

Recorded evaluation configuration: {"d":"n"}. This record does not assert independent reproduction.

### GraphDR-shuffled

Cumulative Regret R(T): 650.2. Result ID: 6a421a69fbc373592c870d61.

[Reporting paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.md>)

Recorded evaluation configuration: {"Type":"control"}. This record does not assert independent reproduction.

### JL+LinUCB

Cumulative Regret R(T): 660. Result ID: 6a421a69fbc373592c870d60.

[Reporting paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65801e209131cea0a935.md>)

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

### Random

Cumulative Regret R(T): 677.6. Result ID: 6a421a69fbc373592c870d5f.

[Reporting paper](<https://www.sota2.com/research/paper/6a41fbe6206542251d7cc9e4.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.
