# Federated Class-Incremental Learning on ImageNet Subset 10 incremental tasks

Source: [Federated Class-Incremental Learning on ImageNet Subset 10 incremental tasks](<https://www.sota2.com/research/sota/federated-class-incremental-learning-on-imagenet-subset-10-incremental-tasks>)

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

Data updated: 2026-02-26T05:28:10.961Z

Evaluation of model performance using Federated Class-Incremental Learning methodology on the specific subset of ImageNet defined by '10 incremental tasks'.

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 | Federated Class-Incremental Learning |
| Dataset (including any stated split) | ImageNet Subset 10 incremental tasks |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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/federated-class-incremental-learning-on-imagenet-subset-10-incremental-tasks.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Avg Accuracy](<https://www.sota2.com/research/sota/federated-class-incremental-learning-on-imagenet-subset-10-incremental-tasks.md?metric=Avg%20Accuracy>): Higher is better; unit: PERCENT.

- [Improvement (Delta)](<https://www.sota2.com/research/sota/federated-class-incremental-learning-on-imagenet-subset-10-incremental-tasks.md?metric=Improvement%20(Delta)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GLFC (Ours) | 42.7 | Not recorded | 68fa6c0d1e209131cea31b3c |
| Ours-w/oCGC | 39.3 | Not recorded | 68fa6c0d1e209131cea31b3c |
| DDE + iCaRL + FL | 39 | Not recorded | 68fa6c0d1e209131cea31b3c |
| PODNet + FL | 38.3 | Not recorded | 68fa6c0d1e209131cea31b3c |
| SS-IL + FL | 37.3 | Not recorded | 68fa6c0d1e209131cea31b3c |
| Ours-w/oPRS | 37.3 | Not recorded | 68fa6c0d1e209131cea31b3c |
| GeoDL + iCaRL + FL | 37 | Not recorded | 68fa6c0d1e209131cea31b3c |
| Ours-w/oCRD | 35.3 | Not recorded | 68fa6c0d1e209131cea31b3c |
| BiC + FL | 34 | Not recorded | 68fa6c0d1e209131cea31b3c |
| iCaRL + FL | 32.7 | Not recorded | 68fa6c0d1e209131cea31b3c |

### GLFC (Ours)

Accuracy: 42.7. Result ID: 698ebd65ffc3901de0a03e4b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Implementation](<https://github.com/conditionWang/FCIL>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

### Ours-w/oCGC

Accuracy: 39.3. Result ID: 698ebd65ffc3901de0a03e48.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Implementation](<https://github.com/conditionWang/FCIL>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10,"Ablation":"Without CGC"}. This record does not assert independent reproduction.

### DDE + iCaRL + FL

Accuracy: 39. Result ID: 698ebd65ffc3901de0a03e45.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa696f1e209131cea22081.md>)

[Implementation](<https://github.com/JoyHuYY1412/DDE_CIL>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

### PODNet + FL

Accuracy: 38.3. Result ID: 698ebd65ffc3901de0a03e44.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa686b1e209131cea1bffa.md>)

[Implementation](<https://www.sota2.com/github.com/arthurdouillard/incremental_learning.pytorch>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

### SS-IL + FL

Accuracy: 37.3. Result ID: 698ebd65ffc3901de0a03e47.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd81e209131cea36380.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

### Ours-w/oPRS

Accuracy: 37.3. Result ID: 698ebd65ffc3901de0a03e4a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Implementation](<https://github.com/conditionWang/FCIL>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10,"Ablation":"Without PRS"}. This record does not assert independent reproduction.

### GeoDL + iCaRL + FL

Accuracy: 37. Result ID: 698ebd65ffc3901de0a03e46.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ce1e209131cea243e5.md>)

[Implementation](<https://bit.ly/3adLNub>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

### Ours-w/oCRD

Accuracy: 35.3. Result ID: 698ebd65ffc3901de0a03e49.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Implementation](<https://github.com/conditionWang/FCIL>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10,"Ablation":"Without CRD"}. This record does not assert independent reproduction.

### BiC + FL

Accuracy: 34. Result ID: 698ebd65ffc3901de0a03e43.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65dc1e209131cea0ce26.md>)

[Implementation](<https://github.com/srebuffi/iCaRL>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

### iCaRL + FL

Accuracy: 32.7. Result ID: 698ebd65ffc3901de0a03e42.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c0d1e209131cea31b3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fded4.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Exemplar memory size":"2000","Federated Learning":true,"Number of incremental tasks":10}. This record does not assert independent reproduction.

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