ReLU (Rectified Linear Unit)

A widely used activation function in deep learning that outputs zero for negative values and the input itself for positive values, enabling faster training and reducing vanishing gradients.