Learning Rate Schedules
Inverse Square Root Schedule
Inverse Square Root is a learning rate schedule 1 / $\sqrt{\max\left(n, k\right)}$ where $n$ is the current training iteration and $k$ is the number of warm-up steps. This sets a constant learning rate for the first $k$ steps, then exponentially decays the learning rate until pre-training is over.
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Language Modelling | 98 | 8.98% |
| Question Answering | 61 | 5.59% |
| Decoder | 51 | 4.67% |
| Text Generation | 43 | 3.94% |
| Sentence | 43 | 3.94% |
| Retrieval | 37 | 3.39% |
| Translation | 29 | 2.66% |
| Machine Translation | 25 | 2.29% |
| Natural Language Understanding | 21 | 1.92% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
