The answer
No. It propagates derivative information through the computation graph. That information describes how changes in intermediate quantities and parameters affect the loss.
What is backpropagation, and why is it necessary?
Does backpropagation send the original input backward?
More FAQs in What is backpropagation, and why is it necessary?
No. Backpropagation computes gradients efficiently. Gradient descent is an optimization rule that uses those gradients to choose new parameter values during training.
Ordinary response generation uses forward computations with fixed learned parameters. Backpropagation is normally reserved for training or other explicit procedures that need gradients.
Still have questions?
Browse all our FAQs or reach out to our support team
Want to upskill yourself?
Your learning library
Small steps. Big possibilities.
Find your next course. Build something great.