Cosine Annealing
Introduced by Loshchilov et al. in SGDR: Stochastic Gradient Descent with Warm RestartsCosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before being increased rapidly again. The resetting of the learning rate acts like a simulated restart of the learning process and the re-use of good weights as the starting point of the restart is referred to as a "warm restart" in contrast to a "cold restart" where a new set of small random numbers may be used as a starting point.
$$\eta_{t} = \eta_{min}^{i} + \frac{1}{2}\left(\eta_{max}^{i}-\eta_{min}^{i}\right)\left(1+\cos\left(\frac{T_{cur}}{T_{i}}\pi\right)\right) $$
Where where $\eta_{min}^{i}$ and $ \eta_{max}^{i}$ are ranges for the learning rate, and $T_{cur}$ account for how many epochs have been performed since the last restart.
Text Source: Jason Brownlee
Image Source: Gao Huang
Source:
Latest Papers
Tasks
| TASK | PAPERS | SHARE |
|---|---|---|
| Image Classification | 17 | 38.64% |
| Object Detection | 9 | 20.45% |
| Self-Supervised Image Classification | 2 | 4.55% |
| Real-Time Object Detection | 2 | 4.55% |
| Unsupervised Representation Learning | 1 | 2.27% |
| Image Generation | 1 | 2.27% |
| Domain Adaptation | 1 | 2.27% |
| Unsupervised Domain Adaptation | 1 | 2.27% |
| Lane Detection | 1 | 2.27% |
Usage Over Time
Components
| COMPONENT | TYPE |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |