Machine Learning & AI

Advanced

Positional Encoding

Injects word-order information into a Transformer that otherwise ignores position.

Formula

PE_{(pos,2i)}=\sin\!\left(\dfrac{pos}{10000^{2i/d}}\right)

Variables

posToken position

iDimension index

dModel dimension

Example

Different frequencies encode position uniquely

Did You Know?

Sinusoidal encodings let Transformers generalize to sequence lengths never seen in training.

transformer nlp deep learning

Share this formula

More in Machine Learning & AI

Linear Regression Model

Predicts a continuous value as a weighted sum of input features plus a bias.

Gradient Descent Update

Iteratively moves parameters in the direction that most reduces the loss.

ReLU Activation

Rectified Linear Unit: outputs the input if positive, else zero.

Leaky ReLU

A ReLU variant that lets a small gradient flow for negative inputs.

Tanh Activation

Squashes input to the range (-1, 1); zero-centred activation.

Softmax

Turns a vector of scores into a probability distribution that sums to 1.