Learning Rate is a hyperparameter that controls how big a step gradient descent takes during parameter updates. A high learning rate can speed up training but risk overshooting minima, while a low rate may lead to slow convergence. Tuning the learning rate is critical for stable and efficient training.