Learning Rate Schedules
Slanted Triangular Learning Rates
Introduced by Howard et al. in Universal Language Model Fine-tuning for Text ClassificationSlanted Triangular Learning Rates (STLR) is a learning rate schedule which first linearly increases the learning rate and then linearly decays it, which can be seen in Figure to the right. It is a modification of Triangular Learning Rates, with a short increase and a long decay period.
Source: Universal Language Model Fine-tuning for Text Classification
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| General Classification | 13 | 18.06% |
| Language Modelling | 12 | 16.67% |
| Text Classification | 10 | 13.89% |
| Sentiment Analysis | 9 | 12.50% |
| Hate Speech Detection | 3 | 4.17% |
| Marketing | 2 | 2.78% |
| Transliteration | 2 | 2.78% |
| Language Identification | 2 | 2.78% |
| Pretrained Language Models | 2 | 2.78% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
