Adam is an adaptive learning rate optimization algorithm that utilises both momentum and scaling, combining the benefits of RMSProp and SGD w/th Momentum. The optimizer is designed to be appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients.
The weight updates are performed as:
$$ w_{t} = w_{t-1} - \eta\frac{\hat{m}_{t}}{\sqrt{\hat{v}_{t}} + \epsilon} $$
with
$$ \hat{m}_{t} = \frac{m_{t}}{1-\beta^{t}_{1}} $$
$$ \hat{v}_{t} = \frac{v_{t}}{1-\beta^{t}_{2}} $$
$$ m_{t} = \beta_{1}m_{t-1} + (1-\beta_{1})g_{t} $$
$$ v_{t} = \beta_{2}v_{t-1} + (1-\beta_{2})g_{t}^{2} $$
$ \eta $ is the step size/learning rate, around 1e-3 in the original paper. $ \epsilon $ is a small number, typically 1e-8 or 1e-10, to prevent dividing by zero. $ \beta_{1} $ and $ \beta_{2} $ are forgetting parameters, with typical values 0.9 and 0.999, respectively.
Source: Adam: A Method for Stochastic OptimizationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Language Modelling | 60 | 7.15% |
Retrieval | 42 | 5.01% |
Question Answering | 37 | 4.41% |
Large Language Model | 32 | 3.81% |
Decoder | 24 | 2.86% |
Image Classification | 15 | 1.79% |
Text Generation | 15 | 1.79% |
Sentence | 14 | 1.67% |
Semantic Segmentation | 14 | 1.67% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |