# Image Classification on CIFAR-100 (Directional Accuracy Metrics) Source: [Image Classification on CIFAR-100 (Directional Accuracy Metrics)]() Collection index: [llms.txt]() Data updated: 2026-07-07T09:16:36.160Z Evaluation of standard Image Classification performance on CIFAR-100 using directional accuracy metrics (Dir-0.1, Dir-0.3, Dir-0.6), typically used for assessing directional prediction or uncertainty quantification. 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 | Image Classification | | Dataset (including any stated split) | CIFAR-100 | | Metric | Accuracy (Dir-0.6) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 (Dir-0.6)](): Higher is better; unit: PERCENT. - [Accuracy (Dir-0.3)](): Higher is better; unit: PERCENT. - [Accuracy (Dir-0.1)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Dir-0.6) | Result date | Report paper | | --- | --- | --- | --- | | FedACT | 66.84 | Not recorded | 6a4c880785dba69fef8b267a | | FedAdamW | 65.43 | Not recorded | 6a4c880785dba69fef8b267a | | SCAFFOLD | 64.96 | Not recorded | 6a4c880785dba69fef8b267a | | FedAvg | 64.93 | Not recorded | 6a4c880785dba69fef8b267a | | FedProx | 64.54 | Not recorded | 6a4c880785dba69fef8b267a | | FedLADA | 64.3 | Not recorded | 6a4c880785dba69fef8b267a | | LocalAdamW | 64.23 | Not recorded | 6a4c880785dba69fef8b267a | | LocalAdam | 63.95 | Not recorded | 6a4c880785dba69fef8b267a | | FedAdam | 62.55 | Not recorded | 6a4c880785dba69fef8b267a | ### FedACT Accuracy (Dir-0.6): 66.84. Result ID: 6a4cc3db60b88bb339c95fb0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### FedAdamW Accuracy (Dir-0.6): 65.43. Result ID: 6a4cc3db60b88bb339c95faf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### SCAFFOLD Accuracy (Dir-0.6): 64.96. Result ID: 6a4cc3db60b88bb339c95fa9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### FedAvg Accuracy (Dir-0.6): 64.93. Result ID: 6a4cc3db60b88bb339c95fa8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### FedProx Accuracy (Dir-0.6): 64.54. Result ID: 6a4cc3db60b88bb339c95faa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### FedLADA Accuracy (Dir-0.6): 64.3. Result ID: 6a4cc3db60b88bb339c95fad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### LocalAdamW Accuracy (Dir-0.6): 64.23. Result ID: 6a4cc3db60b88bb339c95fae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### LocalAdam Accuracy (Dir-0.6): 63.95. Result ID: 6a4cc3db60b88bb339c95fab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. ### FedAdam Accuracy (Dir-0.6): 62.55. Result ID: 6a4cc3db60b88bb339c95fac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Communication Rounds":"300"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.